204 SOME SINGULAR THINGS 



nal and somethiies the thoracic segments. They 

 are found only in certain groups, but ajjpear to be 

 a universal characteristic of the earliest stage of the 

 Lycaenidae, and have the appearance of spiracles, 

 only they are ordinarily quite circidar, while spir- 

 acles are generally oval, and they present no open- 

 ing in the centre, but only, as far as I have been 

 able to observe, a simple pit of more delicate struc- 

 ture than the chitinous annidus itself. They are 

 found also in some Hesperidae in their earliest 

 stage and sometimes also throughout life. But for 

 the tenuous structure of the pit in the centre, they 

 would have all the appearance of suppressed spines, 

 and, indeed, the central pit seems sometimes to be 

 wanting, and we have simply a shining lenticle, 

 similar to those which are so common in the Papi- 

 lioninae. But whether they should be looked upon 

 as structures on their way to some use, or as effete 

 structures, degenerated spines so to say, we have no 

 facts at present to show, and an explanation of 

 their purpose is still to be sought. 



The discussion of these odd structures and curi- 

 ous arrangements of parts in caterpillars may al- 

 ready have been extended to too great length, but I 

 should like to draw a moment's attention to two 

 other special points in the structure of caterpillars 



