252 APPENDIX. 



first attempt to separate the species given under Coccus 

 in the tenth edition of the ' Systema Naturae ' of Lin- 

 naeus, was made by Geoffrey, in his ' Histoire Abregee des 

 Insectes,' vol. i (1762), where he placed a part of them 

 under Chermes, and left adonidum and phalaridis, with 

 his new species ulmi, under Coccus. Of these species 

 only phalaridis was given under the genus Coccus by 

 Linnseus in his tenth edition, and is therefore the only 

 species that could be regarded as the type of Coccus so 

 far as Geoffroy is concerned. 



"In 1802, Latreille, in vol. iii, p. 267, of his ' Hist. 

 Nat. Crust. Ins./ established hesperidum as the type of 

 the genus Coccus. I have not been able to find that 

 any of the writers between the appearance of the work 

 of Greoffroy and that of Latreille published anything 

 that would fix the type of Coccus. Leach in 1815, and 

 Samouelle in 1819, adopted cacti as the type, but the 

 statement made by Leach that it e inhabits fruit trees ' 

 makes it quite certain that he had under consideration 

 neither cacti, L., nor the cochineal insect. Samouelle 

 merely copies Leach. Curtis, in his ' British Entomo- 

 logy ' (1838), gives cacti, L., as the type ; but none of 

 these three authors could affect the question, as the 

 type had already been established by Latreille, if not 

 by Geoff roy, as shown above. 



" The phalaridis of Linnaeus was so obscure an insect 

 that the author himself could not determine whether 

 it was a Coccus, an Aphis, or a Chermes. Fonscolombe, 

 in describing his Coccus radicum graminis ( c Ann. Soc. 

 Ent. Fr.,' iii, 212, 1834), gave the synonymy as follows : 

 ' Phalaridis (?), Linn., Fab., non C. phalaridis, Enc. 

 Meth. nee Geoffr.' Professor Cockerell has suggested 

 the idea that the phalaridis of Geoff roy was possibly 

 not the same species as the Linnaaan insect, which is 

 precisely the same idea that Fonscolombe had, as 

 shown by his synonymy. Since it is probably impos- 

 sible to prove that Geoffrey had any of the Linnsean 

 species of the tenth edition in his restricted genus, the 

 only safe ground will be to adopt the type established 



