rNTROmTCTTOX. 21 



no greater, have longer mandibles oi (jiiitc a (liliereiit iKitterii, 

 fig. 4 rj-i. iSueii speeimens lonn (Hiitc a distinct j)luise, 

 uneonneeted by any intermediates with the j)rogressive series. 

 They are alwaj-s of full size and, unhke the rest, show 

 practically no variation. 



A very striking examj)le of a Lucanid witii these two distinct 

 phases \s('nlrode.^ iiratus, (Plate XX, figs. H-\ I ), which is found 

 in lunnbers in the Malay Peninsula. It is ratlier a small insect, 

 males varying from 18 to 27 mm. in lengtli, exclusive of the 

 mandibles. It is uni{|ue in its genus for its beautiful metallic 

 colouring and also for tlie fact that, from the smaUest to the 

 largest-sized specimens, tlic maiuHbles show extremely little 

 progressive develo])ment. In a very small male (fig. 8) they 

 are very tiny, less tlian half as long as the head, l)ut as we ])ass 

 to larger and larger slx^eimens we tind only very slight develop- 

 ment, and a specimen of the largest size known may have them 

 only a little more than half the length of the head. But 

 other male specimens occur with very highly-dev^eloped jaws 

 which, as the figure (fig. 10) sliows, bear no resemblance at 

 all to those of the ordinary jjhase. Together they form 

 almost a perfect circle and are toothed internally in a very 

 curious and elegant manner. Again, only large males of this 

 phase are found and no sort of transition a[)i)ears to exist. 



The genus Calcodcs contains many cases of the occurrence, 

 side by side, of the two ])hases, one inconstant and the other 

 constant, the specimens of the latter being usually less numerous 

 than those of the former. The conunon Indian Calcodcs siva, 

 shown in Plate II, tig. 2, is a good cxamjjle. The males have 

 usually short, stout jaws, fitiely toothed at the inner edge 

 (figs. 2, a, h, r), but .some have them long and slender (fig. 2 d), 

 and between the two forms no links are found. The Briti.sh 

 Museum collection contains thirty-three short-jawed males, 

 ranging from the smallest to the largest size, as well as eleven 

 long-jawed ones, all of large size. The same genus is repre- 

 sented in Ceylon by Calcode-n carinatus (Plate II, tig. 3), which 

 is not unconnnon there, and I have seen about fifty males of 

 all sizes of the varial^le ])hase, tigs. 3 a-c, of which the mandibles 

 show a gradual progress from the short, broad Priodont form 

 of the smallest to a narrow, sU-nder form, with a lobe at the 

 base and a pointed branch before the middle. Together 

 with the fifty exam^jles of this variable j)hasc were taken 

 thirteen specimens of an isolated phase, fig. 3 d. in these 

 the mandibles are much longer, the basal lobe and the pointed 

 branch are both absent and, instead of them, there is a forked 

 branch beyond the middle which is not found in any specimen 

 of the variable phase. In Culcodes cuvcra, dcU's.serti and 

 other species of this geiuis two exactly similar j)hascs are 

 foimd, the nmubeis of each bearing a similar prt>portiou to 



