524 W. L. TOWER, 



by Verson (1890), Gonin (1894), Comstock & Needham (1899), 

 Needham (1900), and Mercer (1900). I have observed this in young 

 larvae of L. decemlineafa, Hippodamia ISpmictata and Epilachna 

 horealis, but in Chrysohothris femorata, Phymatodes variabilis and 

 allied forms no such concentric arrangement was found. In those 

 species in which this "rosette" arrangement of cells occurs it is 

 soon lost and the fundament appears as an elliptical area of 

 polygonal cells , which is constantly increasing in thickness. In 

 L. decemlineata^ the cells composing the wing fundament are not 

 distinguishable from those of the surrounding hypodermis excepting 

 that they are more numerous within a given area. However, this 

 same condition is found in other parts of the larval hypodermis and 

 these do not develop into any larval or imaginai organ. The thicken- 

 ing of the fundament continues gradually during the second instar 

 (L. decemlineata) , so that there is eventually formed a marked 

 elliptical plate of cells occupying the entire breadth of each of the 

 last two thoracic segments. This condition I have observed in L. 

 decemUneata (PI. 14, Figs. 1 and 2, PI. 15, Figs. 14 and 16), Coc- 

 cinella bipunctata L., Hippodamia ISpmictata L., Epilachna borealis 

 Fab., Phymatodes variabilis Fab. (PL 14, Fig. 5; PI. 17, Fig. 26), 

 Buprestis rufipes Oliv., Chrysobothris femorata Fab., and Carabus 

 sp.?. In C. bipunctata, H. ISpunctata and E. borealis the disc is 

 shorter than in the other species mentioned, being more nearly cir- 

 cular in outline. In these Coccinellidae it is also more sharply marked 

 off from the surrounding hypodermis than in the other beetles 

 examined, but only rarely have I found it as sharply differentiated as 

 is figured by Comstock & Needham (1899) for H. Idpunctata. In 

 most of the beetles examined there was a gradual transition from the 

 disc to the surrounding hypodermis, such as I have figured for L. decem- 

 Uneata (PI. 15, Figs. 14 and 16), or for P. variabilis (PI. 17, 

 Fig. 26). 



It has been shown by Dewitz and others that these imaginai 

 discs are not circular, but are oval or elliptical in outline, and occupy 

 the larger part of the side of the wing bearing segments. Thus 

 Dewitz (1881) has shown that in TricJwstegia varia Koll. the first 

 stage of the wing fundament is a thickened elliptical area occupying 

 nearly the entire breadth of the segment, and Verson (1890) has 

 shown that in Bombyx niori the wing fundament is an elliptical 

 "rosette" of cells placed where the spiracles of the last two segments 

 ought to be. 



