( 474 ) 



force ns to treat the Sumatraii form not as a separate sjiecios. North Indian 

 cloanthus occur in several broods ; the iudivichials of tlie s])ring and those of the 

 summer broods are in the extent of the markings rather obviously different, and 

 this proves that the species is in pattern easily modified by the transmuting 

 factors ; the extreme individuals of the spring brood difl'er in the extent of the 

 markings more from the extreme examples of the summer brood than certain Indian 

 individuals do from certain Sumatran ones. Further, the differences of certain 

 Chinese specimens from Indian individuals are quantitatively greater than those 

 between Indian and Sumatran examples. Hence it is correct to accept Mr. 

 Rothschild's opinion * and to treat the three insects in question as three subspecies 

 of P. cloniitliiis. 



The valve (f. 149) of P. cloantJius is, as in nil the iillied species, rather small: 

 iit the ajies it is deeply sinuate. The siuns divides the apical third of the valve 

 in a smaller dorsal lobe (fi) and a larger ventral lobe (a). The ventral edge of the 

 valve is angulate (/'), and from this angle to tlie tip of the ventral lobe densely 

 l)eset with irregular rows of thin and sharp teeth. The internal sheath of the valve 

 is raised into a distinct fold (e), which begins ventrally at the base— in f. 149 

 and ISO the right-hand side is the ventral side of the valve — runs for a short 

 distance along the ventral margin of the valve, turns in a rather even curve round 

 towards the dorsal side, forms a subdorsal tooth (e), and then traverses longitudinally 

 the dorsal lobe, being liere raised into a short dentate ridge {d), the " dorsal ridge '" 

 which stands almost perj)endicular upon the ])laue of the lobe leaning very feebly 

 over ventrally. In f. 149 and 1.50 the subdorsal tooth («■) is visilile almost in 

 its entire length, because it is bent over apically, while the dorsal ridge aj)pears 

 mtich less high than it really is. 



The variation of the species according to locality relates especially to tlie form 

 of the valve, the length of tlie sulidorsal tooth, and the form of the dorsal ridge. 



a. P. cloanthus cloantliu.s from Kulu to the Shan States, at higher elevations : 

 f. 149, 151, 152. 



The sinus of the valve is J mm. deep. The ventri-dorsal fold (<;) is slightly 

 curved. The subdorsal tooth (c) is high, simple, and when seen from the dorsal 

 side, as in f. lol and 152, reaches so far that its tij) ajipcars to be above the 

 dorsal ridge. The length of the tooth is variabh; : the lower extreme met with by 

 ns is re])resented by f. 1 52 (Shan States) ; in a second individual from the Shan 

 States the tooth is nearly as long as in f. 151, which is taken from a Sikkim 

 specimen and represents tl\e usual form uf the tootli. As both the Chinese and 

 Sumatran subspecies have the tooth generally considerably shorter, the mountainous 

 regions of Siam, Tenasserim, and Malacca on the one side, and of Upper Tonkin on 

 the other, will most probably yiehl (if inliabitcd by cloanthus) individuals more 

 often, or even constantly, intermeiliatc between the three subspecies in respect to 

 the length of the tootli. 



Tlie dorsal ridge (f. 151, 152) of cloanthus cloanthus is a little longer than 

 high, and is in all our si)ecimens rather strongly denticulate. 



The individuals of the spring brood are not different in the valvf and Iiarpe 

 friini the indiviihi;ils of the later broods. 



b. P. cloanthus cli/mcnus from Western and Central China ; f. 153. 



In tlie outline of the valve all the inilividuals examined agree with sj)ecimens 



• Nov. Zooi.. p. 44:. (18!!.-.). 



