12 ILLINOIS BIOLOGICAL MONOGRAPHS [12 



It is intended that the figures, descriptions, and definitions shall 

 make every statement in this paper absolutely definite and that they 

 shall be so clear that a novice may be able to make a complete description 

 of a caterpillar without the possibility of confusion as to his meaning. 

 New species and new instars should be described in a manner which 

 will make like specimens recognizable in the future without repeating the 

 breeding. Such complete descriptions in the published records are very 

 few at the present time. 



HISTORICAL 



The first worker to find and describe a primary plan of the setae of 

 caterpillars was Wilhelm Midler (1886), in a paper on the Nymphalidae. 

 After discussing the arrangement in the first stage of these butterfly 

 larvae, he says : ' ' We find in the first stage, on the abdomen of all forms 

 considered, certain setae. Tho of many different forms their constant 

 arrangement shows them to be homologous. We call these setae pri- 

 mary. * * Doubt as to the similarity between the mesothorax and 

 metathorax and the abdomen seems to be out of the question." In an 

 appendix, brief notes on the chaetotaxy of the larvae of other families 

 of Lepidoptera are given. Careful comparison is made between the 

 Saturnioidea and the Sphingidae. 



Dr. Harrison G. Dyar did not find this paper and in 1894 again 

 discussed the subject as new and numbered the primary setae. So far 

 as I know, this author makes the first suggestion that the position 

 of the setae be used in classification. His observations included 

 a few representatives of many families and his phylogeny as 

 based on them is remarkably good. He also numbers the metathoracic 

 setae but makes no attempt to show that the seta bearing a certain num- 

 ber on the abdomen is homologous with one bearing the same number 

 on the thoras. He himself realized that the metathoracic "tubercles," 

 i a-f-i b and ii a-j^b (Fig. 11), were not the homotypes of the abdominal 

 setae i and ii (Fig. 12), and in 1901 made a definite statement to that 

 effect. This discussion, therefore, can not be considered a contribution 

 to the study of homotypy. 



In another paper the following year the same author reported 

 observations on the first stages of many larvae, finding that they dif- 

 fered considerably from following stages. By these observations he 

 established the primary arrangement of the setae on the abdomen and 

 demonstrated its uniformity throughout a great part of the order. A 

 few months later, in "A Classification of Lepidoptera on Larval Char- 



