June, 1916.] 



Forbes : Caterpillar Homologies. 



139 



the thorax. Again I believe, though this time mainly without evi- 

 dence on either side, that the setae of Jugatae and Frenatae are homol- 

 ogous, but differ a little in position. So I take the anterior lateral, 

 epsilon, of the JugatJe, to be eta of the Frenatse, and delta of the 

 Jugatae to be epsilon of the Frenatse. 



As to homologies with the abdomen so far as number of setae 

 goes the comparison is close, only that there are two subprimaries on 

 the thorax. Unfortunately the presence and location of the wing 

 and spiracular rudiments makes the homology impossible. Above the 

 spiracle, on the abdomen we have only iii and iiia, while above the 

 wing-rudiment, which must be well above the spiracular level, we 

 have all except eta of the Frenatse on the thorax, and that is far 

 above the actual thoracic spiracular rudiment, which lies on the level 

 of vi (pi). Eta is anterior, and free from the wing, but there can 

 hardly have been extensive migration of kappa and theta as they are 

 closely associated (above) with the wing-base, especially the primary 

 kappa. We have then epsilon, rho, kappa, theta and possibly eta (if 

 it is epsilon of the Jugatas) to compare with the two setae iii and iii a 

 of the abdomen. Under the conditions any comparison is a guess, 

 and Fracker's one that iii a is epsilon, and iii is rho, is as likely as 

 any. Below the spiracle, we have nothing to represent iv, v and vi, as 



Summary. 



11a 



iib 



absent 



absent 



absent 



vii (vi Dyar) 



setae on coxa 



viii 



ix 



delta 

 rho 

 theta 

 kappa 

 (epsilon ?) 



Level of dorso-ventral muscles 



epsilon 



rho 



theta 



kappa 



eta 



absent 

 absent 

 absent 



Level of wings and spiracles 



nu, pi 



sigma 

 tau 



nu, pi 



Sigma 

 omega 



VI 



vii a 

 viic 

 viii 

 ix 



viib 



epsilon 

 rho 



kappa 



eta 



mu 



nu, pi 



tau 



sigma 



omega 



