272 [December, 



liand, whether any were mere weed visitors. Incidentally, also, some idea 

 has been gained as to the species present in West and North Wales, and 

 their relative abundance. 



In compiling these notes no special attention was paid to any par- 

 ticular type of cultivation, etc., notes being made wherever and whenever 

 opportunity offered, whetlier ni field, garden, or on hillside or woodland, 

 so as to try to get as complete a view as possible of the total activities 

 of the species. It must be meiitioned, however, that the mountains 

 were less worked than the lowlands and foothills. Records included 

 llower visits hj Apis wellifica, Aniliopliora pilipes^ and a few by various 

 Fsiiliyri (recorded thus and not specifically). Records of other Apidae 

 or of Golletidae and Andrenidae were not made owing partly to lack of 

 time to extend the scope, but chlefiy owing to lack of knowledge of the 

 species ; although numbers of these are undoubtedly of importance as 

 pollinators of fruit, etc. 



The data were mostly obtained from the following localities : — 

 Crosswood, Cardiganshire, a few ; Aberystwyth district (per Mr. T. A. 

 Ste])henson) ; Aberdovey, a few (for which I am indebted to Mr. C. H. 

 Mortimer) ; Aberhosan, near Machynlleth ; the four North Wales counties, 

 from Pwllheli to near Chester, with a preponderance from the vicinity of 

 Bangor and Bethesda ; about Ludlow (Salop) and Leominster (Hereford- 

 shire), a few; the vicinity of Liverpool (Altcar, Formby, etc.), a few ; 

 North Lancashire (Ulverstone, Bai-row, and the Islands of Roa, Piel, and 

 Foulney), a few records. The majority of records, however, relate to 

 West and North Wales ; others are chiefly in the nature of controls 

 and agree with the Welsh records in every way. The records contained 

 in Knuth's "Handbook of Flower Pollination" and other literature 

 have been consulted, but the data found are not incorporated in this 

 account. 



Records of queens, workers, and males visiting flowers were included 

 under the one heading, although the earlier records were, of course, 

 those of queens, wjiile any males recorded chiefly came under July- 

 September. 



It should be noted that very considerable variations in size occur 

 among all the castes of the various species ; Graenicher, for instance, 

 records variation between 9 and 17 mm. in body-length and from 

 5-11 mm. in tongue-length in the American species, H. pennsylvanicus 

 (s\?e Bull. Wisconsin Nat. Hist. Soc. vol. viii, no. 4, 1910). Such differ- 

 ences will be reflected in corresponding variability in the flowers visited 

 in some instances. 



