AGRIUS CONVOLVULI. 379 



appearances, and that whilst the first-named are almost of necessity 

 immigrants, but may be the progeny of May-June immigrants, the last- 

 named also may consist either of immigrant or native-born individuals 

 in different years. As a matter of fact one can only assume that 

 under the most favourable circumstances two native-bred broods can 

 appear in the greater part of the Palaearctic region, and that these would 

 be in August-September and October-November, and then only in years 

 when immigrant $ s reached us in May-June (a most unusual occur- 

 rence). These two latter (Augustand October) broods occur in Roumania 

 (Caradja), and Fritsch gives dates extending over many years, 

 for Austro-Hungary, from May 16th to November 5th. Two broods 

 appear to have been pretty regularly noted in Malta, and North 

 Africa, viz., May-June and September-October, with an intermediate 

 July-August one frequently developed, but detailed observations 

 and actual dates are much needed. Fletcher gives May 26th, 1902, 

 for Malta, Lucas has recorded imagines in May at Algiers, Seriziat 

 has noted examples in June at Collo, and adds the remark that 

 the September emergence is much more abundant and continued 

 into October. Occasionally, as we have stated, a July flight is 

 observed in Britain, and we note that Zapater records the species 

 as occurring in July at Ateca, and we have also in addition, July, 1896, 

 at light, at Aix-les-Bains (Agassiz), from July 14th to September 10th, 

 1893, at light at Berne (Hiltbold), very common in June- July in 

 Transcaucasia (Romanoff), in mid-July and August at Gorki (Ball), 

 most abundant in July at Sicily (Mina-Palumbo), &c, whilst the 

 general statement of Fritsch that imagines occur pretty continuously in 

 Austro-Hungary from May 16th to November 5th, could be easily 

 matched by taking the earliest and latest recorded dates in Britain 

 over a long series of years ; the fact is, much detailed observation 

 is needed as to the dates of appearance of this species in all 

 countries, with a careful weighing of the probabilities as to whether 

 the recorded specimens are really immigrants or native-born. The 

 conditions governing the appearance of the species in the same 

 year in various parts of the Palaearctic area may be very different; 

 in 1898 they must have been exceedingly alike, and the simultaneous 

 abundance of the species in August and September of that year 

 in all countries across the Euro-Asiatic continent from Britain to 

 Japan is perhaps noteworthy. Of its time of appearance, south 

 of the Equator, we know little ; specimens in the British Museum 

 collection were taken by Marshall at Umtali in December, 1900, 

 and at Salisbury, in Mashonaland, in March, 1896, whilst in New 

 Zealand, Hudson notes the imagines as being most abundant in 

 November and December (anted, p. 336), and so on. So important 

 do we consider the unravelling of the details of the natural history 

 of this species that we have no hesitation in giving the following 

 long list of records of the capture of this species in Britain, trusting 

 that additions in future years will give sufficient materials for safer 

 generalisations than those in which we can at present indulge. These 

 are : August 28th, 1841, several August, 1859, at Hull (Young), August 

 1 2th, 1846, at Mailing Deanery, near Lewes (Merrifield), September 

 3rd, 1847, in Arran (Thomson), September 27th, 1853, at Bridg- 

 water (Clark), September 2nd, 1856, an imago at Peckham (Siggs), 

 September 8th, 1856, at Mortlake (Edon), September 9th, 1856, 



