Swarms were still quite widespread in all three major Regions in early 1946, but spring breeding was on a restricted scale in both the Western Region (where it was confined to SW Morocco, NW Algeria and some localities of northern Sahara) and the Eastern Region, where it was restricted to southern Iran. Some spring-generation swarms were nevertheless formed in NW Africa, and reached the summer breeding areas to the south of the Sahara. The few swarms produced in Iran moved through Pakistan into India, where they spread as far east and south-east as Uttar and Madhya Pradesh, apparently without returning to the usual summer breeding areas. In contrast to the other Regions, quite extensive spring breeding occurred in the northern part of the Central Region, where it took place on a medium scale over the northern half of the Arabian Peninsula, extended into Jordan, Syria and Iraq, and gave rise to a considerable number of swarms. To the south of the Gulf of Aden it took place on a moderate scale over southern Somali Republic, eastern Kenya and northern Tanzania, but gave rise to quite a large number of swarms which spread widely over East African countries.

There was almost negligible gregarious breeding during the summer monsoon in the Eastern Region, where either no swarms or only limited swarms were produced. In the Western Region the reported summer breeding took place on a somewhat restricted scale in Senegal, Mali and western Chad, but it appears probable that breeding was under-reported and that it extended into Niger Republic, where pink swarms appeared in September. It would appear that considerable swarming populations were produced, for in October-November several large swarms invaded Morocco and northern Algeria and spread to Tunisia.

In the Central Region the invasion of the summer breeding areas in NE Africa took place in May and June, and in July numerous swarms were reported in the Sudan, Eritrea and NE Ethiopia; other swarms reached SW Arabia and spread on to the Somali Peninsula. There was widespread and locally heavy summer breeding throughout the belt from west Chad to SW Arabia; considerable swarming populations were produced, and moved to the Red Sea coastal areas and western Arabia by the end of the year. Further south in the Central Region, most of the `Long Rains' (spring generation) swarms produced in East Africa remained in north Tanzania till August and in Kenya till September, when they bred in the north-western part of the latter in Samburu and Turkana; others may have moved north to the northern part of Somali Peninsula, where they became mixed with spring swarms from Arabia.

The widespread summer breeding in the Central Region was followed by notable winter 1946-47 breeding, which developed on both sides of the Red Sea, was widespread over Somali Peninsula, and extended into northern Kenya and NE Tanzania. Most of the egg-deposits in the latter countries failed to hatch, but some swarms were produced in the more northerly areas (I).

There was no reported gregarious breeding in spring 1947 in the Eastern Region, which continued apparently free of any swarming populations. In the Western Region there was breeding of moderate extent in Morocco, Algeria and western Tunisia and some swarms (reported as numerous in Morocco) were formed in all these countries, and emigrated to the summer areas to the south of the Sahara by the end of July. In the northern part of the Central Region some limited spring breeding, which nevertheless gave rise to some new swarms, occurred on the northern part of the Arabian Peninsula and extended into Jordan, Israel, southern Syria and Iraq. Further south, however, some spring breeding took place in Asir, while Long Rains breeding of moderate extent, but giving rise to considerable swarming populations, developed in NE Ethiopia and Ogaden, and in southern Somali Republic and northern Kenya.

During summer 1947 there were two reports of isolated swarms of unknown origin in SE West Pakistan and in northern India. There was, however, no gregarious monsoon breeding anywhere in the Eastern Region, and no records of any swarms of the summer generation until late in the year, when a thin immature swarm was seen on the Mekran coast in Pakistan, and a swarmlet was reported in SE Iran. In the Western Region, summer 1947 breeding took place, on a somewhat restricted scale, in Niger Republic and in Mauritania and Senegal. Some swarms were, nevertheless, produced, and were reported moving through Mauritania and Spanish Sahara and into SW Morocco, in the last quarter of the year.

In the Central Region spring swarms from Arabia moved into the Sudan and countries bordering the southern Red Sea and Gulf of Aden, where they augmented the quite numerous swarms produced there during spring on the Long Rains. In the Sudan the summer rains were late in starting and there were no layings till the second half of August; hatchings were reported in many localities throughout the Sudan summer breeding belt, but were said to have been all controlled, so that the whole country became free of infestations after September. Further east there was some breeding on the western lowlands and central plateau in Eritrea and quite heavy breeding in the Harar Province, with some extension into the western part of the northern Somali Republic, while SW Arabia remained free. Most of the hopper infestations in Eritrea were destroyed by control or by birds, but there were some escapes in Harar. On the whole, in spite of breeding on a medium scale in the Sudan, and on a more restricted scale in the eastern section of the summer breeding areas, only limited summer-generation swarms were produced in the Central Region. There were, however, a number of swarms of the previous Long Rains generation surviving through summer in Ethiopia and on the northern part of the Somali Peninsula. These, together with the few swarms produced in the summer, were widely distributed during the last quarter of the year, when isolated swarms were reported from the Dessie escarpment, Tigrai, Harar and Borana Provinces and further east in northern Somali Republic and moving south through Ogaden and Mudugh. Owing to the failure of the Short Rains, however, there was no winter breeding anywhere on the Peninsula. Further north, the Red Sea coast of the Sudan remained clear, while most of the very limited layings on the Eritrean coast failed to hatch. Similarly, there was no winter breeding on the Red Sea coast of Arabia, which remained clear until one or two small swarms were reported on the Saudi Tihama in December. Thus no new swarms were produced in the 1947-48 winter season and only limited residual swarming populations survived in the Central Region into early 1948 (1).

1st and 2nd quarters (Maps 59-60)
Western Region Following the somewhat restricted summer 1947 breeding in the Western Region, several swarms reached the SW borders of Morocco in November. In January 1948 they invaded the Souss Valley, and subsequently spread through the western part of Morocco west of 5°W. Laying on a restricted scale occurred there between February and May, with hoppers hatching between March and mid-June. The breeding gave rise to several swarms which formed between June and mid-July; these began to leave Morocco in a southerly direction in June, and had all emigrated by the end of July. The rest of NW Africa remained free of infestations throughout the spring of 1948 (1).

Further south a few swarms were reported in February and March in, respectively, northern and SW Mauritania, while in June there were reports of one or two swarms and of some restricted breeding in the southern part of the country (1).

Central Region In Arabia very heavy and widespread rains fell on the Saudi Tihama from Lith to Hali in late November 1947, and resulted in floods which were said to have been the heaviest for twenty years. There were further intermittent light showers over the area in December to February 1948, with more showers falling over the Wadi Doga and Wadi Qanuna areas in April.

Following reports of a few loose pink swarms on the Tihama at the end of 1947 (see above), a small pink swarm and some scattered pink locusts were seen at, respectively, Muwaih (to NE of Mecca) and over Kamaran Island in the second half of January, and a swarm was reported from Medina in early February. Only solitary locusts were, however, found on the Tihama between Jedda and Hali during a reconnaissance carried out between late January and mid-February. The following description of subsequent events is based on accounts by G. Popov (Popov 1948 in (1) ).

In mid-March scattered mature adults and some hoppers were found in several localities between Lith and Hali, with highest densities per 1,000 sq. metres varying from 5 near Wadi Doga to 12 near Wadi Qanuna; the adults sampled in Wadi Qanuna, where locusts were found throughout its length from Haleifa to Qunfida, were solitariform. By the end of March, when the vegetation in the coastal areas was drying out, the population in Wadi Qanuna ranged from mature adults to hatchlings, and through all instars to fledglings. Most of the hoppers were in the fourth and fifth instars and concentrated in the stands of green Heliotropium growing along the edges of dukhn (Pennisetum typhoideum) cultivations. The first and second instars were invariably green but the older stages were showing a progressively increasing dark pattern, and about 25% of the older hoppers were typically gregaricolor. The density of hoppers in the stands of green Heliotropium continued to increase with the further drying-out of vegetation outside the wadis, and on 12th April a marching band of mainly gregaricolor fifth-instar hoppers, measuring about 140 x 180 m. was found on the bank of Wadi Haleifa and controlled. In the meantime the adults, 70% of which consisted by the end of March of newly fledged locusts, were moving into cultivations and forming concentrations in the greener stands of dukhn or along the edges of the dukhn plots. Other concentrations of hoppers and fledglings were formed on Acacia bushes scattered over the sandy plains covered by withering annuals.

In late April there was an increase of young adults in the area, and in early May a loose swarm was seen near the edge of Wadi Qanuna and was thinned out by control. At the same time a loose thin swarm was found to the south of Qunfida in Wadi Shafa. The morphometrics of the young adults sampled in Qanuna area were solitariform-to-transient.

The gradual concentration and gregarisation of hoppers as a result of differential rates of drying-out of vegetation were observed also in the Shaq and Wadi Doga areas, where, by the end of March the hoppers had reached the third to fifth instars and were concentrated mainly on green Heliotropium bushes; most of them were green but some were developing black markings. On 18th April, when most of the annuals in the area had dried out, some gregaricolor late-instar hoppers and several small loose swarmlets of young adults from 0.5 to one sq. km. in size were found in the Pennisetum typhoideum cultivations and in the foothills between the Doga and Shaq Yamani wadis. Between the 20th and 27th April the swarmlets were seen performing local flights, moving in a general southerly direction and concentrating in the dukhn cultivations. Most of them finally merged into one compact swarm 3-4 sq. km. in extent, which remained in the Wadi Doga area till the end of April; the morphometrics of the locusts in this swarm were found to be transient to gregariform. In addition, another small swarm was observed moving southwards over Wadi Doga. It should be mentioned that the parents and the early stages of the small-scale gregarious infestations in the foothills were not seen; by analogy with observed events in the lower Wadi Doga and in Wadi Qanuna it appears possible that they arose through gradual concentration of scattered progeny of scattered parents; on the other hand, they might have resulted from concentrated layings by small swarms, since swarming populations were seen on the Tihama as recently as the preceding February.

Apart from day flights by groups and swarms, there was a considerable amount of night-flying by scattered locusts, and near Wadi Qanuna up to 60 of them per minute could be seen flying against the moon on 20th April (Graham 1948, in (1) ).

In addition to the newly formed swarms observed in the breeding areas, a pink swarm was seen in the Lith area between 28th April and 3rd May, when it moved away to the south. After that date there were no further reports from Saudi Tihama or the interior, and the fate of the swarms is not known; some of them may have moved over to Eritrea, while others might have moved to SW Arabia where there were unconfirmed reports in June of groups at Ahwar.

In the African part of the Central Region, following the failure of the 1947-48 Short Rains breeding on the Somali Peninsula some mature and immature swarms (most probably the remnants of swarming populations formed in the Central Region in spring and summer of 1947) were reported in the winter of 1947-48 in Ethiopia, along the Dessie Escarpment, in the Railway area, and further south in the Rift Valley; in February 1948 there were only a few reports from Rift Valley, Danakil and Harar Plateau, and the last report from Ethiopia (apart from Eritrea) referred to a small mature swarm seen near the Somali Republic border to the west of Borama in early March. The only report over this period from Somali Republic was of a few scattered adults seen in the Galkayu area in January 1948 (1).

Further north there was some breeding by scattered adults on the Red Sea coast of Eritrea, where transient hoppers were found in January in Wadi Falcat and at Mersa Taclai; some scattered adults, again described as transient, were noted in Wadi Falcat in June (1).

In the Sudan, control operations were carried out in April against loose bands found on the Red Sea coast over an area of c. 1 sq. km. at Khor Egaiet to the north of Port Sudan, and in May against later hatchlings to the north of Port Sudan and at Eit and in the Halibai area (1).

Still further north in the African part of the Central Region, yellow locusts laid in March 1948 in Mendil in Harug el Asued in Fezzan, but the hoppers which hatched from these layings were said to have all died from the sun and hot wind (Guichard 1955). Another informant, however, stated that Harug "was full of red and yellow locusts at the time" (Guichard l.c.); this suggests that the parent generation might have been swarming, and that their breeding may have given rise to further gregarious locusts.

Finally, after heavy rains over northern Egypt during the 1947-48 season, Schistocerca hoppers appeared during June in the barley fields at Sollum, Sawyet el Tufaya and Negeila areas on the Mediterranean coast (1).

There were no reports of swarms arising as the result of spring breeding in any of the above areas. Yet in the second half of June Eritrea was invaded by several swarms, consisting mostly of immature locusts but with a small proportion mature, and reported between 16th and 30th on the central plateau and the coastal plain. During the same period, an immature swarm was seen in the Sudan at Tokar, and there were several reports (believed to have referred to the same mature swarm) from Shendi district in Northern Province (1).

Eastern Region In SE Arabia nil reports were returned during this period from the Muscat State. In the interior of Trucial Oman, 22 specimens of ph. solitaria Desert Locusts were collected in April at Buraimi by Thesiger, who thought that locusts were quite numerous in the area; the collection which he brought included, however, a number of grasshoppers, which may have been confused with the locusts (1).

In Iran, where a two-hectare swarm was seen flying over Wisham and Lashar in Mekran in early December 1947, two small swarms were seen in the SE part of the country in early February; in mid-May two small mixed swarms appeared in the Bampur district, and there were reports of hopper infestations over some 30 hectares on the coastal plain to the north of Chahbar (1) ; other hopper infestations occurred during the season in the Bampur depression (Popov 1953).

Following reports of very few scattered locusts or none in Pakistan and India in the early part of the year, the populations in southern Baluchistan began to rise after widespread rains in March. By late April there was a concentration of laying immigrants in the Dasht Valley in SW Mekran, where the maximum estimated density rose to 1,000 per sq. km., and where gregarious hoppers were observed in May over an area of 21 sq. km. near Suntsar, and were controlled. There were further notable rains in the coastal area and the interior of Baluchistan in June, and scattered populations continued to be reported, reaching an estimated maximum of 800 per sq. km. at Gwadar.

In the meantime scattered populations remained very low in India until the second half of June, when they began to rise owing to immigration and reached an estimated maximum of about 3,500 per sq. km. in jodhpur (1).

3rd and 4th quarters (Maps 61-62)
Western Region In Morocco, the last of the young swarms emigrated in a southerly direction on the 20th July. In the same month an immature swarm was reported flying over Tindouf in Algerian Sahara, and in mid-July several loose and smallish swarms were seen in Mauritania in Chinguetti, Tidjikja and Moujeria areas; it is possible that some of these were Anacridium. There was little rain in Mauritania except in the extreme south till August, and all the swarms which had invaded the country disappeared without any recorded breeding, and Bruneau de Miré (1952) considered that the invading swarms may have dissociated owing to drought. This suggestion is consistent with the complete absence of any reports of swarms from other West African countries to the south of the Sahara, and with the observed dissociation of swarms in similar circumstances during the drought in NW India in the early summer of 1963 (p. 64). There were no other reports of swarms in the Western Region for the rest of 1948, except for a swarm in Gao area in Mali in December; in view of the records of Anacridium swarms and breeding in Mauritania, in the preceding July and September, and of Anacridium swarms in Nigeria in January 1949 and in Mali in June 1949, it appears highly probable that this swarm was of this species, which was then swarming in many parts of West Africa. Between August and October only scattered solitariform locusts were found in Mauritania by Bruneau de Miré (1952), in several localities between Atar and Aioun el Atrouss; these were apparently very sparse, except in local concentrations in Tamga Massif and to south of Tichitt.

Central Region On the Mediterranean coast of Egypt, hopper infestations continued through July and into August, when late-instar hoppers, and adults ranging from solitaria to gregaria, were found in Sollum, Sawyet el Terfiya and Negeila, mixed with hoppers and adults of Anacridium aegyptium and Locusta sp. The adults were apparently aggregating into swarmlets which were drifting eastwards and were reported by newspapers from Sollum on 9th, Sidi Barrani on 12th, and Mersa Matruh on 14th August; it is possible, however, that these swarmlets were in fact of Anacridium which is known to swarm in these areas; on 18th August there were two newspaper reports of swarms in Upper Egypt near Aswan and in Wadi Half a, and again there is no certainty that these were not of Anacridium(1).

In Eritrea some of the swarms which invaded the country in June remained till early July, when there were three more reports of immature swarms on the central plateau between Kheren and Nacfa. Similarly there were a few more reports of swarms in the Northern and Khartoum Provinces of the Sudan between 1st-22nd July, and on the 12th a swarm was reported from Darfur; in view of a report of Anacridium melanorhodon hoppers in the same area in the following October, it is probable that this swarm was of this species. No summer breeding by Desert Locusts was observed in either country, and there were no further reports of swarms after July until November, when some "flying locusts" (possibly a loose swarmlet) were seen on the Red Sea coast of the Sudan at Halaib. Finally, on 6th December there was an unconfirmed report of a swarm on the Eritrean coast near Badda.

Further south in the Central Region some swarming populations may have persisted in the second half of the year in central Ethiopia, where there were unconfirmed reports of immature swarms on, respectively, 9th October and 8th December in the Rift Valley to north and north-east of Adama (1).

There were no reports of any locust activity in Arabia during the second half of 1948.

Eastern Region In Pakistan, following a further increase among scattered locusts in southern Baluchistan (to an estimated maximum of 3,200 per sq. km. recorded in the Kulanch Valley by early July), there was scattered solitary breeding in July and August in the Kalanch and Gwadar areas. It was heaviest in the Sardasht sector, where initially solitary hoppers formed many bands, which were controlled. Some breeding, which required control operations, was reported also in the Lasbela State. In the meantime the adult population declined over most of Baluchistan owing to dispersal of adults and control of hoppers, but at Pasni it rose to an estimated maximum of 9,200 per sq. km. in early September.

In Rajasthan in India, where the maximum estimated density of adults fell off in July to 1,000 per sq. km. and in August to only 300 per sq. km. there was solitary breeding in jodhpur and in western Bikaner; in one locality in the latter the density of second and third instar hoppers reached 100-300 per 0.1 sq. metre, and control operations were carried out. Solitary hoppers in various stages continued to be present in Bikaner in September where the resultant population of adults rose to an estimated maximum of about 15,000 per sq. km.

In October the population declined to 200 per sq. km. in Bikaner, and rose to 1,000 per sq. km. in Nokh, but the maximum in Rajasthan declined to 100 per sq. km. in November, and no locusts were observed there in December. Further west the population was stated to have been nil in Pakistan in the first half of October, except at Pasni, where it reached 4,200 per sq. km. declining by the middle of November to 1,200 per sq. km. After that date only a few locusts were to be seen in Baluchistan, where they were observed in small numbers at Pasni and in Turbat.

There were no reports of any locust activity in the second half of the year in Iran, and no information on the locust situation in SE Arabia (1).

Between 25th and 28th October 1948 a tropical cyclone from the Arabian Sea moved into SE Arabia and produced heavy rain which fell over a wide area extending from Dhofar and Dakaka to Mugshin and northwards through Wadi Ain, Buraimi and Sabkhat Mutti to the Trucial coast (Thesiger 1950-51; Rainey 1962, 1965). According to Rainey the windfield was affected by the cyclone over an area with a radius of a thousand kilometres, and the air which was brought into the cyclone could have collected airborne locusts throughout the region extending from Eritrea to Baluchistan. It was estimated further by Rainey that within the affected area the windfield must have exhibited a degree of convergence sufficient to increase greatly the area density of randomly orientated fliers in a period of a few days.

A Schistocerca outbreak in South Africa
Following unusually good rains in the second half of the 1947-48 season the populations of the Desert Locust built up to large numbers in the first half of 1948 in Gordonia district of the Cape Province, in Karasburg and Aroab districts and the Namib Desert of South-West Africa, and in western Bechuanaland bordering the Nosop River. Over 400 incipient hopper bands were destroyed in Gordonia and Karasburg districts, but scattered hoppers reached the adult stage in all the affected areas. The fliers tended to move south and west, and further control measures against swarms were undertaken in Karasburg district and in the irrigation settlements of Vioolsdrift on the Orange river. Large populations of adults were still to be found in all the above areas, and in Namaqualand, in the second half of the year (Neveling 1949).

1st and 2nd quarters (Maps 63-64)
Eastern Region In SE Arabia, according to local reports, a loose mixed swarmlet visited Mudafin, in the Liqtan Valley to SE of Dubai, Trucial Oman, in the first week in February, and on 10th February W. Thesiger (who was travelling in the interior of Oman through the areas which had received rainfall during the October 1948 cyclone) found loose concentrations of late-instar hoppers and pink and yellow adults in Wadi Ain running from the western slopes of Jabal Akhdar into the eastern sands of Rub Al Khali.

During a survey on 20th-22nd February, scattered pink, grey and yellow individuals were found in Trucial Oman coastal areas, giving estimated densities of up to 2,600 per sq. km.; in the last days of February some mixed swarms appeared on the Batina coast of Muscat, and after two days moved off to the interior, and in early March a mixed swarm appeared between Dubai and Abu Dhabi.

In the second week of March there were several reports of mixed and mature swarms from the Rumaiya area in the Liqtan Valley SE of Dubai, and in the Ras el Khaima area the populations of scattered locusts rose, after a shower, to 4,600 per sq. km. between 18th and 23rd March. Breeding by these scattered locusts gave rise to hopper populations which ranged from scattered to transient and gregarious; in the Liqtan Valley there was patchy gregarious breeding over an area of some 150 sq. km. with hoppers hatching from the 29th March; in addition there were reports in mid-April of hopper bands in the area between Abu Dhabi and Buraimi, and of heavy breeding in Buraimi and in adjoining areas to the north-west.

In most areas fledging began in early May, and between 9th and 14th several young swarmlets and swarms were seen in the Ras el Khaima area. On 12th May the Bedu reported large concentrations of swarms, stretching over some 50 km. to NW of Buraimi, and there were rumours that young swarms left the Liqtan Valley in early May.

The specimens of the parent generation invading Trucial Oman, collected in late February and early March in the Liqtan Valley, had transient E/F ratios, whereas the specimens of the new generation taken in May in several parts of Trucial Oman were morphometrically transient, or transient to gregariform, but contained a proportion of solitariforms (data from Desert Locust Survey, Nairobi, collections).

In Iran only isolated solitariform locusts were seen in several sections of the coastal areas between February and May, but their numbers were somewhat higher near the Pakistan border in the Dashti district, where breeding began in March. In April a group or a swarmlet was reported not far from that area, in Bawaridan (1).

The population of scattered locusts remained very low in both Pakistan and India in the first quarter of 1949, when it varied in coastal Mekran between 30 and 45 per sq. km. During April there was an increase in the population in the Dashti Valley in SW Baluchistan where it rose to 800 per sq. km, and scattered locusts appeared all over Baluchistan. In May light breeding described as 'bordering on gregarious' began in Nakkar and Rakshan areas near Panjgur, and extended into Gwash area in Kharan in June, when the population of scattered adults in Chackro in Sind rose to an estimated maximum of 460 per sq. km. The formation of new swarms as the result of Baluchistan breeding was prevented by control operations ( (1); Ahmad 1950).

In India there was a rise in the scattered populations in Rajasthan to an estimated maximum of 300 per sq. km. in late May, and the infiltration of immigrants continued through June and July. The monsoon broke in the second half of June and rainfall was particularly heavy in the Phalodi district of jodhpur, and Kaku Badla area of Bikaner. Maturation set in after the first showers, and layings and hatchings had begun by the end of June ((1); Pruthi 1950).

Western Region There were no reports of Desert Locusts in this Region throughout this period. A swarm of unidentified species reported in Louga area in Senegal in March 1949 was, in all probability, of Anacridium sp., swarms of which were reported in the second half of 1948 and in 1949 in several parts of West Africa (cf. p. 55).

Central Region There were no reports throughout this period of any locust activity in the Chad Republic or in U.A.R., and only solitary locusts were found in January-February on the Red Sea coast of the Sudan and Eritrea. Good rains fell on the Eritrean coast north of 17°N in February, with frequent showers continuing in Karora area in March. Breeding must have begun on the coast in February, for incipient band formation was observed on the northern coast of Eritrea in Karora-Halibai area by mid-March. The parents were solitariform and scattered and although their density gradually increased with time, they showed no tendency towards swarming behaviour (Stower 1959). Hatchings continued till mid-April and all the first instar hoppers were green, but later instars acquired an increasingly transient to gregarious coloration. At Halibai the first filial population was estimated to be some 200 million over an area of about 20 sq. km. of Pennisetum cultivation. The later instars grouped, without forming discrete bands, and began to march, eventually forming a large moving population some 20 km. wide and up to 6 km. deep, in which the density of older hoppers varied from 0.1 to above 25 per square metre. Fledging began in April. The population was thinned by control operations. By 20th May all the remaining hoppers had fledged and the young adults were morphometrically "cool congregans" (Stower 1959; Stower, Davies & Jones 1960; (1) ).

The building-up of hopper populations and their gregarisation extended across the border on to the southern coast of the Sudan, where hatchings near Karora and in several khors to the north must have begun in February or March. Small bands of early instars were seen in early April, by the middle of which older hoppers were forming loose marching bands, one of which marched on a 12 km. front. At the same time locusts were fledging and forming groups, one of which measured 450 x 450 metres, and on 23rd April a small swarm was seen flying at a point some 30 km. NN W of Karora.

During May the adults moved inland both in Eritrea, where they were described as dispersing among the foothills, and in the Sudan where considerable numbers were reported from Wadi Safra and Borgeig in the Northern Province. There may have been further merging of groups or swarmlets, for between 7-12th June small swarms were seen on the Eritrean plateau near Nacfa and Gher, a swarmlet appeared in June at Sabderat and on 17th June a swarm flew over the coast at Arafali.

There were no reports from other parts of Ethiopia, and only solitary locusts and limited solitary breeding occurred on the Gulf of Aden coast of the Somali Republic in February to May.

In the Arabian part of the Central Region there was only solitary breeding on the Saudi Arabian Tihama in January-March, with fledglings disappearing by April, but from the first half of May swarms began to be reported from western and southern Arabia. Thus in early May locusts appeared at Badiya, in Wadi Mugshin, and on 10th May several mixed swarms entered Wadi Hadhramout from the north between Naqra and Heenim. At about the same time (7th May) a swarm was seen at Sabiya, on the Saudi Tihama. Some of the swarms invading Hadhramout remained in Eastern Aden, where they were reported at Hurghada and Risit in the second half of May; others moved west reaching Beihan by 12th and Dhala and Ladat on 22nd. The locusts were mostly yellow, but with a proportion of pink, yellow and grey individuals; in the last ten days of May and in early June mostly small swarms were repeatedly reported from the Aden-Yemeni border. In the meantime there was an influx of scattered locusts on the Saudi Tihama, a swarmlet was seen at Taif, and in early June a large swarm moved north between Abha and Taif, while in Yemen small mature swarms reached Taizz from the east in late May.

The locusts which invaded Wadi Mugshin oviposited in May and may have given rise to some gregarious progeny, which remained uncontrolled. Further west in south Arabia swarms laid in May in East Aden in wadis Minwah and Bin Ali, and in Lodar and Shuqra areas in June; all the resultant hopper infestations in these areas were reported to have been controlled in June and July. In addition heavy breeding was reported in June-July in southern Yemen to the west of Beihan, and over at least 260 sq. km. between Marit and Sana'a, while in June some unconfirmed breeding was rumoured at Mocha. The swarms which were formed in the interior of Yemen had emigrated by the 2nd July (1).

The morphometrics of the stragglers from swarms invading West Aden in May, which were collected in June, ranged from transient to gregarious, while the samples of the progeny taken at Shuqra, where breeding was reported as dissocians, were transient to solitariform. The adults of the second generation collected in July in Wadi Bin Ain in East Aden were transient to solitariform (data from Desert Locust Survey, Nairobi, collections).

It will be seen that the breeding which took place in the late spring and early summer in the southern and SW part of the Arabian Peninsula was widespread, and it is possible that it gave rise to considerable swarming populations which may have contributed substantially to the resurgence of the major plague which developed by the end of the year (see pp. 59-60).

3rd and 4th quarters (Maps 65-66)
Eastern Region The immigration of scattered adults into Sind and Rajasthan continued in July, when their maximum estimated density reached 7,400 per sq. km. in Kakku area of Bikaner (1). A sample of 268 specimens taken in that area in mid-July was found to be morphometrically intermediate between solitariform and gregariform, rather closer to solitariform, but with 94 per cent of individuals with 6 eye-stripes, characteristic of the gregarious phase (Misra 1953). In the meantime layings and hatchings continued in many parts of Rajasthan, and gregarious hoppers were reported from Kakku Badla and Hadda areas of Bikaner and Ajasar-Rohena area of northern jodhpur. There was some concentration of immigrant adults in all these areas, which consisted of flat gravelly tracks interspersed with low sand-dunes and shallow cultivated basins. After the first rains the flat tracks became overgrown by dense stands of Cenchrus biflorus and Andropogon halipense, whereas the tops of the dunes remained partially devoid of vegetation and provided clear patches for concentrated layings. The hoppers observed at the time of hatching were solitaricolor, and became grouped on moving into adjoining cultivations; the older hoppers acquired gregarious coloration, and the fledglings, which began to appear from late July were pink and mostly gregariform (Gurdas Singh 1952).

Layings by scattered adults continued through August, and in September when countless populations and dense concentrations of adults were observed in the Bikaner (Pruthi 1950a; (1) ). These may have been formed of rapidly maturing locusts of the first monsoon generation (though it is possible that some of them, e.g., a swarm laying in Sind in early September, might have consisted of concentrated immigrants, or of a mixture of two overlapping generations).

In September and October there were layings by swarms in Jaisalmer and Bikaner in India and in Thar Parkar in Sind and Khairpur and Bahawalpur states in Pakistan, followed by serious hopper infestations in October-November and production of a number of new swarms in November-December (Pruthi 1950, Ahmad 1950; (1)). A westward migration of swarms began in mid-November, when swarms were reported in SW Pakistani Baluchistan.

Swarms from the east arrived in Iranian Baluchistan and Kerman from mid-November, and the invasion continued in December, when there were very many reports from southern Iran, where the swarms spread to the west of Bandar Abbas by the middle of the month. Further west, an influx of scattered immature locusts occurred in northern Oman in the second half of November, and in late November-early December scattered locusts and groups were reported from the Batina coast, and some small swarms moved over Sur near Ras el Hadd (1).

Central Region In SW Arabia, following the termination of the spring and early summer breeding (p. 57), an immature swarm and numerous scattered locusts were seen in August-September at Dhala and Ghar Sudan in Aden, and in September a small swarm was reported from the Saudi Tihama near Gizan. Scattered solitary locusts were seen in the Tihama in October-December, while in mid-December scattered mature locusts appeared in the interior of Aden in the Riyan-Mukalla area, and a yellow swarm was reported from Minwah. In the same month clouds of yellow locusts flew over Taizz in Yemen (1).

In the African section of the Central Region, following the apparent dispersal of the swarmlets produced in Karora, only low-density scattered populations were reported in the second half of the year from the coastal areas or the interior of Kassala Province in the Sudan Republic, while in Ethiopia only light dissocians breeding was observed in August and September to the north of Cheren in Eritrea and in some wadis in Danakil, east of Dessie. In November-December a thin population of scattered locusts was found over the coastal plain of SE Sudan and Eritrea with some concentrations at Khor Asrai, Obellet and Edd, and in the last days of December, when mature swarms were appearing in SW Arabia, there was an unconfirmed report of a flight of a swarm over Arafali.

On the south coast of the Gulf of Aden, where only a few scattered solitariform locusts were found between the French Somlaliland border and Las Dureh in October, the population of scattered locusts apparently increased in November, when a 'locust infestation' was reported between Heis and Mait. Towards the end of November there were reports of swarms from Las Khoreh and of laying near Karrin, and on 1st December fishermen reported a mile-square swarm flying over the Gulf of Aden towards Elayu. There were more reports of layings near Karrin and of locusts appearing at different points along the coast in December, when hatchings began at Las Khoreh. The morphometrics of the immigrant locusts were found to be transient approaching gregariform and the males among them were described as markedly yellow (1).

Western Region There were no reports of any locust activity from this region.

1st and 2nd quarters (Maps 67-68)
In the Eastern Region some of the swarms which had remained in Pakistan through the winter spread northwards to northern Baluchistan, SE Afghanistan and Punjab. Spring breeding began from February in Pakistan and from March in Iran, and developed in these countries on a moderate scale between March and June, giving rise to a number of swarms which started moving towards the eastern summer areas in May-June.

Central Region Following the reappearance of the swarms in SW Arabia towards the end of 1949, some gregarious breeding developed in early 1950 in the interior of Aden and in Yemen, and extended on to the Red Sea coastal areas of Saudi Arabia. It is of interest to note that a case of incipient aggregation in solitary fledglings was observed in January on the Tihama before the locality was reached by swarms in early February, after which their subsequent fate was masked by the presence of swarming locusts (Ellis & Ashall 1957). In the second quarter of the year breeding continued in Aden and Yemen, and spread further north along the Tihama and into Nafud in northern Saudi Arabia. Numerous swarms were formed on the Arabian Peninsula, and began to move south-west, south and east in May and June.

Only solitary or limited transient breeding developed during the spring on the coastal areas of the Sudan and Eritrea, but further south the swarms which had invaded the northern coast of Somali Peninsula in late 1949 gave rise to quite heavy hopper infestations in the coastal areas in early 1950; these were augmented when more mature swarms from Arabia reached the southern coast of the Gulf of Aden and NE Ethiopia in March. In the northern part of the Somali Peninsula most of the escapes from earlier breeding fledged by April, but in April-May further breeding developed in Mijertein, Ogaden and Mudugh, as well as in Danakil, giving rise to more spring-generation swarms.

The invasion by Arabian swarms of the African countries bordering the Red Sea began from May, and in June the immigrant swarms had spread throughout the summer belt in the Sudan and into Chad Republic and further west.

Western Region The only report from the western part of this region (apart from a report of an unidentified swarm, almost certainly of Anacridium melanorhodon in southern Mauritania in late January) referred to some scattered solitary locusts near Tabelbala in NW Sahara in March. In the eastern part of the Region, however, dense bands of hoppers were found in Wadis Waur and Tao, in western Tibesti in north-western Chad, in January (Guichard 1955). Since there had been no records of swarms in that general area for some 18 months, and heavy rains fell over Tibesti in the preceding October-November, the possibility that these represented a local outbreak cannot be excluded (though the interval between the rains and the appearance of dense bands is too brief for a local build-up). There were, however, no reports of resultant swarms, but from June, some of the swarms invading Sudan and Chad apparently spread westward into Niger Republic, where a high-flying swarm was seen at Zinder (1).

3rd and 4th quarters (Maps 69-70)
Eastern Region The invasion of the eastern summer areas by spring-generation swarms from Pakistan Iran and Arabia continued in July, when numerous swarms spread far to the east over northern India. The resultant summer breeding took place on a medium scale, and gave rise to considerable new swarming populations.

In the Central Region the invasion of the Sudan and Ethiopia also continued in July, and between late July and September widespread and heavy breeding developed in the Sudan-Chad section of the breeding belt extending on a moderate scale into northern Ethiopia. Numerous swarms were produced during this breeding, and some of them moved in September northwards from the Sudan, invading U.A.R. on their way to Arabia; others moved eastwards into the Red Sea area, while others again moved out to the west and north-west, invading the Western Region.

At the end of the year some winter breeding developed, on a medium scale, on the Somali Peninsula and around the Red Sea, giving rise to further swarms, and increasing their number in the Central Region.

During this season some scattered populations consisting of solitariform local adults and scattered gregariform stragglers from immigrant swarms (Stower, Davies & Jones 1960) became concentrated into a laying swarm in association with localised rainfall on the Eritrean coast. At the end of November these locusts were present at densities of 1 to 100 per 90 metres at a series of localities along a 90-km. traverse near Massawa. A fortnight later they were found, at densities of 1 to 600 per 90 m., only along a 13-km. stretch through an area which had, in the meantime, received some rains. Four days later the locusts in the area were found to be concentrated into a small swarm covering an area of about 1 sq. km., in which counts of 400-600 per 90 m. of traverse were recorded (observations by Stower, quoted in Rainey 1951).

Western Region It appears that a number of unreported swarms reached the Western Region during the spread of Arabian swarms along the African summer breeding belt, for some hopper infestations developed in Niger Republic in the last quarter of the year, and in Mauritania between September and November. These gave rise to a few swarms, which joined the numerous swarms which moved into the Western Region from east Chad and Sudan, and spread far and wide across the Sahara and into NW Africa, from Morocco to northern Sahara and Libya, by November (1).

As can be seen from Fig. 2 and Map 71, spring breeding occurred in all the Regions, though on a somewhat moderate scale in the Western Region, and swarms produced during spring reached the summer breeding areas to the south of the Sahara, as well as those in the Central Region and in the East. The extent of the summer breeding (Map 72) was on a medium scale in the Eastern Region, and resulted in the production of considerable populations of swarms. In the Central Region it was largely restricted to the eastern half of the Sudan summer area, without extending either into Chad or to southern Arabia, but the supply of swarms in the Region was supplemented by further swarms produced in the following winter on the Somali Peninsula and on the African coast of the Red Sea (Map 73). On the other hand the extent of the summer breeding was notably restricted in the Western Region, where nearly all the available swarming populations became concentrated, at the time of the breeding, into SW Mauritania and Senegal, and where the production of summer-generation swarms was greatly reduced by an effective control campaign ((1); Rainey 1965a).

In the Western Region the limited swarming populations arising from summer breeding in Senegal and Mauritania and the very restricted summer breeding in Niger Republic (Map 72) began to reach NW Africa in October 1960. In the following four months an active campaign was carried out in Morocco against these swarms, contributing effectively to their final disappearance by mid-February, and before they matured and bred (Rainey 1965a). The complete absence of breeding in NW Africa broke the swarming cycle in the Western Region, which has since then remained free of any significant swarming populations until the time of writing, viz., late 1965 (1).

It is of interest to note that after the Niger Republic had been free of swarms since November 1960, and the whole of the Western Region since March 1961, the only swarming populations reported in the Region in summer 1961 were some mature swarms appearing in Termet Massif in August and in Tamesna in September (Map 75); the hoppers which hatched in the Termet in August were controlled. The Tamesna swarm laid in September but most of its progeny either failed to hatch or died soon after hatching owing to drought, while the swarm itself was said to have dispersed though it may have moved into Algerian Sahara where a mature swarm was reported in October (1). The populations found in Tamesna in October consisted of breeding locusts with solitariform-to-transient morphometrics, occurring over an area of 1,000 hectares in stands of Schouwia at the density of 20-30 per 100 paces. The hoppers to which they gave rise were initially solitaricolor, but gradually concentrated into groups of up to 10-20 per square metre, and became increasingly gregaricolor with age. This population was thinned out by control (Germeaux, unpublished).

In the Central Region spring breeding was quite widespread through the central and northern parts of the spring breeding area, though restricted on the Somali Peninsula to its northern part (Map 74). The following summer breeding was of only moderate extent, but swarming populations in the Region were again supplemented as the result of winter breeding on the Somali Peninsula and on both sides of the Red Sea (Maps 75 and 76).

In the Eastern Region the recorded spring breeding was on a restricted scale (Map 74), possibly because a proportion of summer 1960 generation swarms produced in the Eastern Region did not emigrate to the spring areas, but spread far and wide over eastern and peninsular India (Map 73) where most of them probably perished. In spite of the limited spring breeding in the Region, quite numerous swarms (originating in Iran and in the Central Region, and possibly including local swarms from the 1960 generation which might have survived in India and been drawn back into the summer breeding area) gave rise to summer 1961 breeding, which was of medium extent (Map 75). This in turn gave rise to considerable swarming populations, which spread through Pakistan, Afghanistan and Iran, and reached Arabia (Map 76) (1).

No swarming populations were observed during 1962 in the Western Region except for an isolated record of a swarm in Tunisia in January; there was another isolated record in that month from the western borders of the Central Region in Ennedi. In the Central Region there was quite widespread spring breeding in the northern and central parts of the spring breeding areas, but again no breeding in the southern part of the Somali Peninsula (Map 77), while in the Eastern Region spring breeding was exceptionally widespread and heavy, extending into northern Iran, Afghanistan, north Pakistan and NW India. Large populations of swarms were produced in both the Eastern and Central Regions. In contrast to the usual spread of spring swarms from Arabia and countries to the north towards the south-west, south and east, all the numerous swarms formed in these areas moved, with the swarms produced in the Eastern Region, eastwards and south-eastwards into the summer breeding areas of Pakistan and India, which experienced an exceptionally heavy invasion.

Owing to lack of immigration of Arabian swarms into Sudan, and successful control of spring breeding on the northern Somali Peninsula and in Ethiopia, only small-scale swarming populations remained in the Central Region by summer 1962, when some restricted breeding took place in Ethiopia and SW Arabia (Map 78). The following winter breeding on the northern Somali Peninsula was negligible (Map 79), and only very limited swarming populations persisted in the Central Region by the end of 1962.

In the Eastern Region the heavy invasion of the summer areas led to exceptionally widespread, heavy and protracted breeding in Pakistan and India, during which at least two, and in some cases possibly three, successive generations were produced between July and November, but only relatively few swarms were finally produced. The cause of the striking failure of this very heavy breeding is not clear, though a contributory factor may have been the introduction of a more effective method of aircraft spraying of hopper populations during the last stages of the campaign.

Such summer swarms as were produced spread at the end of the year either westwards to southern Iran and eastern Arabia or northwards into northern Pakistan to produce more swarms there in the following spring (p. 63) (1).

1st and 2nd quarters (Maps 80-81)
Western Region There were no reports of any Desert Locust activity during this period from this Region.

Central Region Following the unconfirmed December 1962 reports, which suggested that some of the summer swarms from the Eastern Region or from Africa, may have spread to Arabia, there were further unconfirmed reports of a swarm in Asir in January, and of a copulating swarm to the north of the Nafud in February; it is possible that the Asir swarm consisted of dragonflies, swarms of which were observed at that time in the area. Only isolated locusts were reported in Saudi or in SW Arabia during the rest of this period, and there were no reports of any gregarious hoppers. Exceptional rains fell over large parts of southern Arabia in association with a tropical cyclone which crossed the southern coast from the Arabian Sea in late May, but on this occasion no special locust developments followed.

In the African section of the Central Region a small swarm, covering 2 km., estimated to contain some 60 million solitariform locusts, and thought to have immigrated by night from across the sea, appeared in February on the Eritrean coast to the north of Massawa. The swarm dispersed over an area of 30 sq. km. and gave rise to hopper bands, but both the parents and the hoppers were eliminated by aerial spraying with only a few locusts fledging in the area in early April. Further north along the coast groups of hoppers, of all instars but mostly fifth, containing some gregaricolor individuals and marching on disturbance, and large groups of young adults were found in late March-early April over some 50 km. of the Sudan coast near Port Sudan, where practically no locusts were left by mid-May.

Further south in the Central Region there were some unconfirmed reports of isolated swarms near Diredawa and Jigjiga in the Harar Province of Ethiopia in January, and in the second quarter of the year isolated and in most cases mature swarms were reported from near Dessye, from the Danakil lowlands, from Lake Margherita, and from Yavello in Sidamo-Borana; during the same period there was an unconfirmed report of a swarm between Hargeisa and Berbera in Somali Republic and of a swarm near Mogadiscio. While some of these reports may have referred to the same swarm, there is no doubt that a few swarms survived in the African section of the Central Region throughout the first half of 1963. There were no reports of gregarious breeding, but a small immature swarm seen NE of Addis Ababa in June might have been the spring-generation progeny of these residual populations.

Eastern Region During January there were three reports of swarms in SE Khuzistan in Iran, and in March a swarm was seen on the Batina coast in Oman; there were, however, no reports of breeding in the country. The swarming population reported in West Pakistan in December moved in January-February in a general northward direction and appeared in several districts of Northern Baluchistan, North-West Frontier Province, and northern Punjab. Maturation set in early, after widespread rains in mid-February, and between the second half of February and early April layings took place in Lyallpur, Campbellpore, Rawalpindi, Dera Ismael Khan and Peshawar Divisions.

Hatchings began in April and continued till early May, by which time hopper infestations were reported from 125 villages. Most of the infestations were controlled, but in early June scattered young adults were reported from Peshawar, and some small swarms from Quetta Rawalpindi, Kohat and Dera Ismael Khan. These swarms moved in a general southward direction and passed through Mianwali, Muzaffargarh and Multan Divisions. Most of these locusts must have reached Rajasthan, where populations prior to May were very low, and where, by the end of June, scattered gregariform locusts appeared in most western districts, and a swarm of grey locusts was seen at Jaisalmer (1).

3rd and 4th quarters (Maps 82-83)
Western Region There were no reports from the Western Region until November, when mature locusts with solitariform to transient morphometrics were discovered scattered at the density of 7-8 per 100 paces over some 30 sq. km. in a stand of Schouwia purpurea near In Abangharit in Tamesna NW Niger Republic (i.e., in the area where some degree of gregarisation of scattered locusts may have occurred in 1952 and 1961, pp. 28 and 62). Most of these locusts were found to have disappeared by the end of December.

Central Region In Arabia there were no reports of any Desert Locust activity until mid-November, when a few red locusts were seen flying high over Taif; such flight by day would be characteristic of swarming Desert Locusts, and their colour suggests that they had a gregarious history. The only other record referred to some dozens of locusts which were blown by a dust storm into a hotel at Jedda on the evening of 22nd December; their colour was described as grey, with a tinge of pink, and it is not known if they were Desert Locusts or Anacridium m. arabafrum, which may occur in groups on the Arabian Red Sea coast.

Following a report of a mature swarm at Antelale in Central Danakil in Ethiopia, small groups of hoppers were found in that area in mid-July. During the same month there was an unconfirmed report of a swarm near Borama in the extreme west of northern Somali Republic, where some scattered pink locusts were seen on the coastal plain near Berbera.

There were four more unconfirmed reports of isolated swarms in Ethiopia in 1963; viz., a swarm at Beda in Wallo in mid-September and swarms near Diredawa, at Awash and at Mello in October. None of these could be substantiated despite thorough investigations, which indicated that they were, in all probability, false. In the last two months of the year only isolated locusts were reported from the Guban in Somali Republic and from the coastal plain. in Ethiopia.

Eastern Region Following the first incursion of spring-generation locusts into Rajasthan in late June, there were two further reports from Jaisalmer in July of swarms consisting of grey and pink locusts; at the same time gregariform individuals continued to be found among the extremely sparse scattered populations in many parts of Rajasthan. Finally a red swarm was seen in late July in Khairpur district, and there were unconfirmed reports of groups of locusts from Khipro and Mirpur Khas in Sind, Pakistan. The last report relating to spring generation refers to a group of yellow locusts seen at Chor in the Thar Parkar, Pakistan in September.

Only sparse solitary breeding was observed in Pakistan and India on the monsoon, which was late and did not break in the desert areas till late July.

Only scattered locusts at low densities were seen in NW India in the second half of the year, where their highest estimated densities reached 4,500 per sq. km. in one locality in Bikaner in October, when the adults of the new generation were making their appearance; their numbers fell off in November, and the locusts virtually disappeared in December.

In Pakistan there were no further reports of any locust activity after September until early December, when a "concentration" of dark pink locusts, covering 0.2 x 0.4 km., was reported from Digri in Sind, but could not be found by subsequent air and ground surveys. It is not clear from the report whether it referred to a cohesive swarm or to numerous scattered locusts encountered over the indicated areas. At the end of the month there was an unconfirmed report of a one-sq.-km. swarm at Larkana. No immigration of locusts into Tran was detected and only isolated locusts were seen (in December 1963) in Muscat and Trucial Oman (1).

1st and 2nd quarters (Maps 84-85)
Western Region There is no information on the events in this region until the second quarter of the year. During May low-density hoppers were present over an area of several tens of hectares in Wadi Tamanrasset in southern Algerian Sahara to SE of Silet; in June some apparently scattered ("quasi-isolés") locusts, which may have originated from the breeding in southern Sahara, were seen flying over Silet oasis and settled for some days in Wadi Amdad, while further north a light flight (which might have been a small loose swarmlet) was seen at Aoulef (1). Further east some adult locusts were seen near Bardai in Tibesti in NW Chad in late April-May, apparently moving off in a diffuse flight; it is not clear whether this report refers to scattered locusts or a group. Finally, in Libya some groups or small bands of hoppers and young adults (apparently mixed with other acridids) were controlled in June to SW of Hon.

It is of interest to note that an isolated hopper and two extreme solitariform adults of Schistocerca were found by K. Guichard on Fuerteventura Island during his visit to the Canaries in May 1964; the last occasion on which swarms of Desert Locusts reached the island was in October 1958.

Central Region The only reports over this period from Arabia refer to a few pink locusts near Sakaka, northern Nefud, in January, and to a few yellowish-green (i.e., solitaricolor) individuals on the Yemen Tihama in June. It must be noted that immature solitary Desert Locusts are sometimes of a peachy-pink colour (cf. plate 23 in Rao 1960), and the description of the locusts as `pink' does not necessarily mean that they had a gregarious history.

All but one report from the African sections of this Region referred to very low-density populations of scattered locusts, observed on the coasts of northern Somali Republic in January and of Eritrea in January-February and April, and in Addis Ababa and Ogaden in April. Some very low-density mature Desert Locusts, with extreme solitariform morphometrics, were seen in late March and again in mid-April, over an area of 1 sq. km. watered by the overspill from a nearby artesian well and overgrown by scattered bushes of Hyoscyamus muticus, to the south of Baris in the Kharga Oasis in U.A.R. Some scattered locusts were found in this area when it was first surveyed in October 1963, though no swarming populations have been observed in U.A.R. since June 1961.

In late March there was an unconfirmed report of a swarm in Danakil, but a thorough investigation in that area suggested that the report may have arisen through confusion of grasshoppers with Desert Locusts.

Eastern Region Some isolated Desert Locusts were seen on the southern coastal plain of Iran in January-February, and in the second half of June a quantity of hoppers and two swarmlets (one of them one hectare in extent) were reported from the Jiroft area of Kerman; it is not known whether these swarmlets were formed by fledglings from hopper bands or by aggregation of scattered individuals after they fledged. No spring breeding was observed in SE Arabia, where a loose group apparently passed through Abu Dhabi area in Trucial Oman in early April.

In the eastern part of the Eastern Region only extremely low-density scattered populations were observed in Rajasthan during spring. In Pakistan, however, spring breeding developed from April in the Turbat area of Mekran where very small bands or group! of solitaricolor hoppers were found scattered over some 40 sq. km., with the main infestation restricted to 8 sq. km. of cultivations; these infestations were controlled by mid-May, but it was thought possible that other small-scale hopper infestations occurred in other parts of Mekran. During May scattered adults were apparently widespread in Mekran and Kharan, while in early June the populations seen in Panjgur were described as heavy; these populations were found to have left the area by the end of the month. It would appear from these indications that quite considerable populations may have been produced during spring in SE Iran and SW Mekran.

The immigration into the summer areas began from May (when the estimated maximum densities of scattered locusts in Rajasthan rose to 50 per sq. km. in Jaisalmer) and continued in June (when the estimated maximum rose to 600 per sq. km. in Bikaner), and in all probability into July. Locusts collected in Kolayatji had solitariform-to-transient morphometrics. Following the passage of a cyclonic depression accompanied by abundant rains from the Arabian Sea over Kutch, Thar Parkar and W. Rajasthan, a swarmlet of solitariform-to-transient grey locusts, covering an area of 0.25 sq. km., appeared on 17th June, and subsequently dispersed, in the Sundra area of Barmer district of Rajasthan, where the estimated scattered populations rose to 900-5,400 per sq. km. over an area of 20 sq. km. after the dispersal of the swarmlet, and to a maximum of 9,000 per sq. km. in. early July. As is suggested by Varma & Sharma (1966), this swarmlet may have represented an assemblage of initially scattered! locusts concentrated by convergent air flow associated with the cyclone. It would appear that some layings may have followed the: cyclonic rains, for subsequent data on hopper populations in Barmer and Jaisalmer suggest that laying was likely to have started: in these areas about mid-June (1).

3rd and 4th quarters (Maps 86-87)
Western Region Some scattered solitaricolor locusts, with solitariform-to-transient morphometrics, were found in very small numbers in July-August south of Termet Massif in southern Niger Republic, and in August some very low-density mature solitariform individuals were seen in Tamesna in NW Niger, and a few immature locusts near Boundeit in SE Mauritania; further east some very low-density populations were observed in July to the south of Koro Toro in Chad Republic. It will be seen that in the third quarter low-density scattered populations were apparently widely distributed throughout the summer breeding belt to the south of the Sahara.

In the last quarter of the year scattered locusts at very low densities were found in Tamesna in northern Niger and in several wadis between In Guezzam, Tin Zaouten and Bordj in southern Algerian Sahara, suggesting that northward displacement of low density populations was in progress. In late December 1964 and in January 1965, small diffuse swarms of undetermined species were observed in, respectively, the Tindouf area of NW Algerian Sahara and in Fort Trinquet area in northern Mauritania; it appears highly probable in view of the appearance of Schistocerca hoppers and swarmlets in lower Wadi Draa area in SW Morocco in May-June 1965, that these swarms were, in fact, of the Desert Locust.

Central Region In the western part of this region, in Ennedi, scattered, morphometrically solitariform-to-transient Desert Locusts were found by M. Germeaux in September along a traverse of 80 km. between Fada and Bichaghara, mostly at very low densities, but rising locally to 10-20 individuals per 100 paces; other populations rising to 10 per 100 paces were found over c. 100 sq. km. in Wadi Kika, and lower densities along the c. 100 km. traverse between Wadi Kochili and Largeau. It appears probable that these populations, scattered over very large areas, may have comprised very large numbers of locusts.

In October and November other extensive low-density populations were found in Ennedi in Wadi Ker, on the border of Bichagara Massif, over several thousands of hectares at Mare de Goumeur, over 500 hectares in the Mourdi depression and over several tens of hectares near Yakou; in the last two localities they consisted of late-instar hoppers and newly fledged adults. In early December scattered locusts appeared at light in a number of buildings at Largeau, suggesting that possibly considerable scattered populations were on the move at night.

Further east, in the Sudan Republic, a group of pink and solitariform locusts moved at night over Khartoum in September; in the same month groups of pink locusts were seen not far from the coast at Hiya and Sinkat, and in the Red Sea hills west of Mohammed Qol. In October groups or bands of hoppers and groups of fledglings were controlled in three localities around Bayiunda, in Northern Province, where small groups of adults were seen along the Nile between Korti and Merowe in November.

In Ethiopia and Somali Republic there were reports in July-September of isolated locusts on the plateau in Eritrea, and of a few pink locusts on the coast of the Somali Republic at Zeila; in mid-August there were three reports of isolated immature swarms in the Rift Valley east of Addis Ababa in Ethiopia, and north-west of Borama in the extreme west of northern Somali Republic, but none of these reports could be substantiated in spite of thorough investigations. In September, however, scattered solitariform pink locusts were found, apparently in some numbers, at Sardo and Tendaho in Danakil, and in the middle of the month a small thin swarm of immature locusts was seen from a dhow, flying over the Gulf of Aden, 110 km. SE of Aden, during a period of strong southerly winds.

Towards the end of the year a concentration of mature solitariform locusts, estimated to number over two million bred on the Eritrean coast near the Sudan border at Halibai; hatchings began in late December, with hoppers forming small bands; more bands appeared in the area in January 1965 and were controlled in the following month.

In the southern part of the Central Region there were unconfirmed reports in November of a small swarm near Mogadiscio; in the same month a swarm was seen near Dusa Mareb, while in December scattered locusts were found over some 400 sq. km. in the coastal areas of NW Somali Republic. It appears probable that some small-scale swarming populations did in fact, occur in the southern part of the Central Region, for in January a small swarmlet of 0.5 sq. km. was reported near Mogadiscio and a breeding swarm was rumoured near the Kenya border. No gregarious breeding was, however, found anywhere in the area.

The only reports for the second half of 1964 from western Arabia refer to some low-density scattered adults and green hoppers controlled near Gizan in September, to an unconfirmed record of some scattered fledglings near Taif in October, and to very low-density scattered locusts seen on the Tihama in November and December.

Eastern Region During the summer of 1964 several swarms were produced in the Indo-Pakistani summer breeding areas, and in view of the intrinsic interest of all processes leading to plague upsurges, an attempt is made in the following paragraphs to describe the events leading to their appearance. Much fuller information is available on the developments in India than in Pakistan, but since much of the significant breeding occurred in adjoining areas of the two countries or extended across the international boundary, description of them will be combined in a single narrative. It should be mentioned that in reconstructing the probable sequence of generations from the incomplete available data, the durations of incubation and hopper development were taken to be, respectively, 12-14 days and about seven weeks, i.e., the times these stages usually take in Pakistan and India during the summer season (K. Wardhaugh, unpublished). The shortest time from fledging to sexual maturation is taken to be 2-3 weeks.

There is no information on the level of scattered populations which must have been reaching Sind and Bahawalpur in May and June when they were appearing in Rajasthan. On 14th July, however, about a month after the appearance and dispersal of a swarmlet in the Barmer district (p. 64), loose groups of solitary (presumably solitariform) locusts were seen in the Mithi area, east of Khipro (Sind); these could not be traced subsequently, and may have dispersed, but on 31st July a group or a swarmlet estimated to cover 0.4 hectare was seen near Khipro.

It would appear from these isolated records of the evanescent swarmlet in Barmer and the group(s) in Sind (which might have contained, respectively, a few million and a few hundred thousand individuals), that quite considerable immigrant populations (F₀) had invaded the Indo-Pakistani summer breeding areas; it is clear, moreover, from the distribution and timing of the subsequent scattered breeding that these relative concentrations could have represented only a proportion of the total immigrant populations, which probably built up through repeated immigrations from Mekran and SE Iran.

Reference has already been made to cyclonic rains in Sind and SW Rajasthan in mid-June; in July there were good and widespread rains in Rajasthan extending into Bahawalpur, and scattered rains in Hyderabad and Khairpur, with further widespread rains in NW India and adjoining parts of Pakistan, affecting also Lasbela, during August. As a result, conditions favourable for breeding persisted in the summer breeding areas from mid-June through July and August and locally into September.

As noted above (p. 64), some layings must have started in Barmer and southern Jaisalmer from about mid-June, and in Barmer, where by late July the hopper populations included all instars, the fledging of the first monsoon (F₁) generation apparently began by the end of July; in Jaisalmer, where second to fourth instar hoppers were seen on 23rd July, the fledging must have started in the first half of August. From the protracted simultaneous occurrence of early and late instars in July and August it appears that layings by F₀ adults must have been staggered or possibly performed repeatedly in both areas and in the second half of August there were indications that mixed populations of hoppers were beginning to group. Thus on 17th August, 50 per cent of third instar hoppers at Pochina, in the Maijlar area of southern Jaisalmer, were showing black markings, while on 22nd August control operations were reported against concentrations of third to fifth instar hoppers and fledglings at Phulia, in the same area; in Bolotra area of Barmer by 22nd August mixed populations of first to fifth instar hoppers occurred in groups of up to 100 per Calligonum plant, and the populations of ash-coloured fledglings were described as "countless" (i.e., too many to count). In the Bikaner circle there were no reports of scattered hoppers till late July, but some, apparently small-scale, populations of scattered adults and hoppers were present in Kolayatji area in August where the F_l adults probably began to appear in the second half of the month.

There is no detailed information on the breeding which took place during this period in the summer areas of Pakistan, but first to fifth instar hoppers were present in the last week of August in Cholestan desert of Bahawalpur, while the occurrence in early September of newly fledged adults in parts of Sanghor district of Sind, immediately west of Maijlar area of Jaisalmer, provides strong evidence of undetected breeding in that area in July and August. In addition to this breeding, some hopper infestations were reported during August in a detached area in Uthal, southern Lasbela, where the estimated maximum density of adults was about a thousand per sq. km., and control operations were undertaken against hoppers of all stages and fledglings, in late August and early September, when the estimated maximum density of adults had more than doubled.

It will be seen that the initial breeding by F₀ immigrants took place over a considerable area in Rajasthan and Pakistan. As a result of this breeding, the F₀ generation in the invaded areas must have been continuously augmented, from late July onwards, by the emerging adults of the F₁ generation, this emergence possibly reaching a peak towards the end of August, by which time some of the F₀ adults may have begun to die. The build-up of the adult population found some reflection in the highest estimated numbers reported from different areas; these became higher in most areas than previously: e.g., they rose to about 900 per sq. km. in Sind on 17th July and to over 1,000 per sq. km. in the Bajju area of Bikaner and in the Cholestan desert of Bahawalpur, in the last week of August; at Bajju the locusts consisted of grey solitaricolor immature and maturing individuals, and their estimated density doubled in early September, when they were all reported as mature. It is clear, however, that these estimates provided no real indication of the actual magnitudes of the mixed adult populations which were present within the Sind-Bahawalpur-west Rajasthan area, for between 22nd and 26th August concentrations of mature, solitaricolor (grey) adults, variously described as "loose congregations", "countless over small areas" and, in one case, "a swarmlet", were found breeding at a number of localities in an area of 480 sq. km. between Ramgarh, Ranau and Kishangarh in western Jaisalmer. As mentioned above, it is possible that these concentrations consisted of surviving members of the F₀ generation and rapidly maturing members of the F₁. It is also clear from the recorded instars of hoppers involved in the widespread infestations which developed in September, that relatively concentrated layings must have been taking place intermittently from the first half of August onwards in Khairpur, southern Bahawalpur, western and northern Jaisalmer, western Bikaner, and Barmer. The earlier of these layings would have been by the immigrant F₀ adults (some of which were, in fact, seen in a concentration in the Khipro area of Sind at the end of July, p. 65), while the later layings probably involved increasing numbers of F₁ adults. There is evidence that at least some of these also formed groups or small swarms, for a 0.5 sq. km. yellow swarmlet was seen in the Ramgarh area of western Jaisalmer in mid-September, and a mature swarm was seen in east Mirpur on 4th October. The September hopper infestations (see below), which consisted mainly of mixed instars, were thus most probably made up of both the F₁ and the F₂ monsoon generations.

In the Barmer district (where the first F₁ adults may have begun to fledge from late July) there may have been a brief gap between successive lots of hoppers, but in the last week of September and the first days of October mainly late-instar hoppers and countless fledglings were reported from several localities and control operations were carried out in the Dabli area. Further north, in Jaisalmer, all instars of hoppers were repeatedly reported in the Mohangarh area between late August and early October, but these were, apparently, at low densities only. On the other hand, control operations were carried out in the middle of September against numerous patches and bands of third-to-fifth instar hoppers in the Ranau area of western Jaisalmer, and other hopper infestations developed in the Nokh area of northern Jaisalmer, where by the third week of September grouped first-to-fifth instar hoppers ranged from solitaricolor to gregaricolor, and the estimated densities of grey, bluish and pinkish adults, comprising newly fledged locusts, rose to 17,000 per sq. km. Control operations were carried out in the Nokh area from the second week of September into early October; in the last week of September the fledglings were, locally, too numerous to be counted.

Following the light hopper infestations which occurred in Bikaner in August, further hatchings, at densities requiring control operations, occurred in the Kolayatji area on 12th September, while in the second half of the month mixed-instar hopper bands were being controlled in the Kolayatji, Bikaner, and Pugal areas; those in Kolayatji were described as small-to-medium in size. Fledging began in the last ten days of September (when the densities of adults rose to an estimated value of 7,500 per sq. km. in Kolayatji and Pugal, and they were described as countless in Ranasar), and came to an end in early October.

On the other side of the border, in Pakistan, groups and bands of hoppers of all stages were being controlled in the second half of September in the Islam Garh and Bijnot areas of southern Bahawalpur (in which some infestations persisted till late October). At the same time hoppers were present over some 1000 sq. km. in Khairpur, where gregarisation and band formation was observed over large areas. Control against hoppers and fledglings was in progress till early October, and there were some escapes in both areas. Other infestations occurred in the Khipro desert of Mirpur Khas where hoppers hatched in mid-September, and control against late instars continued till about mid-October.

The adults that escaped control operations (and/or those that arose from breeding in areas where there was no control) formed several swarms; the earliest of these were reported in early October in eastern Khairpur and in Jaisalmer, while in the second week of the month swarms were reported, again in east Khairpur and in Sri Gangangar district, lying to the north and north-east of the infested areas of Bikaner and Bahawalpur. In the second half of the month there were more reports of swarms from Sri Gangangar, Jaisalmer and Sind, as well as Bahawalpur and Bikaner; it is difficult to estimate the size of the swarming population produced, but from the examination of the reports it appears possible that the total area of swarms was of the order of 10-20 sq. km. By the end of October there were indications that the westward movement of swarms had begun, for a swarm was seen at Kalat, in the interior of Baluchistan, and there were some reports from the coastal areas at Pasni. Only one swarm was seen in Rajasthan in November, when some swarms were reported in Pakistan in Lasbela, and in the Turbat area of the interior of Mekran; it appears probable that some of the locusts passed into Iran, and to Oman, where a small group of yellow locusts was seen on the Muscat coast in mid-November, and a few immature locusts were found in the interior of Oman in December.

In December some locusts (which may have been a group or a swarmlet) were again reported on the Mekran coast, between Ormara and Pasni, but there were no further reports of swarms in the Eastern Region through the winter and the following spring (1).