THE CANADIAN ENTOMOLOGIST. 





NOTES ON THE PHYLOGENY OF THE SATURNIANS. 



LY HARRISON G. DYAR, PH. D. , NEW YORK, 



Mr. Grote's remark (Can. Ent., XXVHL, p. 294) that the stinging 

 spines of Hemileuca and Autouieris may have been separately evolved is 

 not in accordance with my views, and I wish to compare his genealogical 

 tree of the Saturnians with the larval characters more at length than was 

 possible in the review of his paper, " Die Saturniiden." I reproduce first 

 his tree : At i there is a dichotomous division, the genera on the right 

 having vein IV._; in the middle of the cell or but slightly moved (generalized); 



those on the left with vein IV2 con- 

 siderably moved toward IVi. It is 

 not proved that this movement of 

 IV2 took place only once in the 

 Saturnians, but it is so assumed, and 

 the construction of the tree depends 

 upon the assumption. 

 Next I present a tree founded on larval characters, using the same 

 generic types. At i is a dichotomous division, the larvje on the right 

 retaining the unpaired tubercle on joint 13 and losing those on the anal 

 plate 



Attacus Saturnia Aglia 



Automeris 



Hemileuca 



Citheronia 



Attacus Saturnia Hemileuca Automeris 



on the left losing th.e 

 unpaired tubercle and retain- 

 ing the pair on the anal plate. 

 At 2 is another division, the 

 two genera above acquiring 

 stinging spines, while Cither- 

 onia remains without them. At 3 the stein of Attacus-Saturnia acquires 

 many haired, reduced tubercles, while Aglia retains the single haired 

 primitive condition and degenerates. 



A comparison shows that these two trees are contradictory, the posi- 

 tion assigned to Aglia and Hemileuca being almost exactly transposed. 

 Yet, if rightly interpreted, there should be no contradiction between larval 

 and imaginal characters. 



If Mr. Grote's tree is correct, Aglia must have reacquired tubercles 

 on the anal plate, because it is derived from the stem of Citheronia after 

 Automeris was thrown off, and neither of these genera possess these 

 tubercles. Likewise, Hemileuca has independently lost these tubercles, 

 unless we suppose that originally they were not present, but were acquired 

 separately by Attacus-Saturnia and Aglia. This can not be, however, as 



