PROSIMULIUM. 15 



TABLE OF PUP^. 



The table of pupae which follows is arranged to include only the 

 North American species associated in this stage with adults. 



1. Respiratory filaments 4 in number 8. bracteatum, p. 39. 



S. johannseni, p. GO. 

 Respiratory filaments more than 4 in number 2 



2. Respiratory filaments 6 in number 3 



Respiratory filaments more than G in number. 



3. Each of the three main filaments divided close to base.iS. meridionale, p. 6G. 

 Each of the three main filaments divided at some distance from base, 



;Si. vcnustum, p. 44. 



4. Respiratory filaments 8 in number 5 



Respiratory filaments more than 8 in number 6 



5. Filaments in pairs 8. piscicidiiim, i).4G. 



Filaments not in pairs (PI. VI, fig. 4) 8. metaUicum, p. 4'.). 



6. Respiratory filaments 9 in number 8. pictipes, p. 57. 



Respiratory filaments 10 or more in number 7 



7. Respiratory filaments 10 in number 8. jenningsi, p. 42. 



Respiratory filaments more than 10 in number 8 



8. Respiratory filaments 16 in number 8. vittatum, p. 55. 



Respiratory filaments at least 22 in number 



9. Respiratory filaments 22 to 25 in number 8. forbesi, i). Qo. 



Respiratory filaments generally more than 40 in number 10 



10. Respiratory filaments at most 48 in number P. pecua>-um, p. 23. 



Respiratory filaments generally 60 or more in number P. hirtipcs, p. 20. 



In addition to the species included in the aboA^e table there are two 

 species described in this stage which have not been linked up with 

 the imago. One of these, from Arizona/ has 8 filaments, and the 

 other, from Santa Cruz Mountains,- has 12. It may be that these 

 species are described in the imago stage and have not yet been 

 associated with the pupa\ 



Since the completion of tlie manuscript of this paper I have had the 

 opportunity of working over the material in the collection of the 

 Illinois State laboratory of natural history which has resulted in the 

 addition of forhesi to the list of species. 



PROSIMULIUM Roubaud. 



Prosimulium was originally proposed as a subgenus by Roubaud.^ 

 The characters which he used for the separation of the group from 

 SlmuUum Latreille were those of the hind tarsus and also the forma- 

 tion of the cocoons, which he considered as possessing a generic 

 value from the fact that they are incomplete, a mere shapeless mass 

 of threads, whereas in SimuJimn the cocoons in the known species 

 are slipper-shaped and .separately formed. It may be that his 



'Trans. Amor. En). Soc, 1893, p. 45. 

 -Bui. tjR. X. Y. Stat.- Mus., lOO::, p. ."S7. 

 »Compt. Rendus .\cad. Sci., Paris, 1906, p. 519. 



