16 ILLINOIS BIOLOGICAL MONOGRAPHS [16 



§tages between radically different conditions. When the tubercles* have 

 been modified to form scoli or verrucae, it often seems impossible to 

 determine whether differences have arisen by coalescence or by the dis- 

 appearance of some of the tubercles. In many cases this is a very real 

 problem, on whose solution depends the correctness of the conception of 

 large groups. 



Another obstacle is the lack of developmental series. In the case of 

 wings, the tracheae which precede the veins in larval and pupal stages 

 plainly show the relation of the adult venation to the generalized type. 

 The single elementary instar which acts as a guide in the study of the 

 chaetotaxy of caterpillars, however, is sometimes so unlike the mature 

 larva in the arrangement of its setae, that it is of little assistance in inter- 

 preting their homology. The change is usually sudden and occurs at 

 molting. Were the recapitulation more often extended over several molts 

 the problem would be easier. 



The third difficulty is one that, up to this time, has prevented prog- 

 ress in this field. Apparently a lepidopterous larva has three or more 

 entirely distinct types of arrangement of the setae (Figs. 7, 8). In only 

 a few species is there any apparent relation between the plan of the pro- 

 thorax and that of the mesothorax, or between the last abdominal segment 

 and any of the others. While this obstacle is not a serious one in classifi- 

 cation, it has prevented thus far the determination of a satisfactory no- 

 menclature and therefore of a working basis. Since Wilhelm Miiller said 

 in 1886 that he found "no support for a comparison of the prothoracic 

 setae with those of the following segments ' ', workers seem to have left the 

 prothorax severely alone. They seem not to have realized that there is 

 evidence not found in the Nymphalidae on which Miiller was working. 



Fortunately there are partial hints and clews to assist in surmount- 

 ing each of these obstacles. Study of varied forms often yields signs of 

 intermediate stages in position. Sometimes the single first stage, our 

 only evidence of the past, unmistakably points the way to an interpre- 

 tation we would not otherwise think of making. Again, the presence of 

 a single unusual seta on a single segment of a generalized form will 

 unite the setal plans of otherwise seriously differing segments. In these 

 ways the gaps are at least partially closed and the problem is taken out 

 of the realm of guesswork and abstraction. 



*The word, "tubercle", is used in this paper as a general term to indicate the 

 location of a seta, or of a definite group of setae, or a process of the body wall 

 bearing such a group. 



