Dammerman: Fauna of the Krakatau Islands. 89 



In the second lot of Danaines I sent to Mr. MOULTON, there was 

 a d" of D. nielanippus melanippus Cr. from Krakatau and Mr. MoULTON 

 wrote me, therefore (17th Aug. 1921):— "This d" is identical with Java 

 examples and 1 fear that on that account my name insularis for the first 

 example you sent me from Krakatau can only indicate an 'aberration' or 

 'form' not a 'geographical race' or 'subspecies' from Krakatau as I thought 

 when describing it." 



About D. javenta krakatauae he wrote me in the same letter:— "On 

 seeing this further series I am now rather doubtful whether krakatauae is 

 a good subspecies; I think it may prove inseparable from the Java form", 

 (D. juventa juventa Cr.). 



Of course as a rule two subspecies of the same species cannot occur 

 together in the same area. This is only possible if the one is a resident 

 form and the other a migrant; or if the two are confined to different 

 habitats, like house rats and country rats. 



But are we perhaps watching the growth of new subspecies on Krakatau ? 

 Is it not possible that "insularis'' and "krakatauae'' are new subspecies 

 in the making, not yet sufficiently fixed, and that the typical "Juventa" 

 and "melanippus" are newly arrived invaders? It is interesting to note in 

 this connection that PlEPERS identified one of the Danaines which JACOBSON 

 collected on Krakatau in 1Q08 as D. ///^/r/////7/7//s /zé'o'^s//7/;ws CR., the Sumatran 

 form of D. nielanippus. If this specimen is in the Leyden Museum, it will 

 be of great interest to see if this form is the true hegesippus; if so 

 Mr. MouLTON may be right in suggesting insularis as a hybrid between 

 hegesippus and melanippus. 



Many other cases of the same kind may come to light when all the 

 material collected on the islands has been worked out. 



V» The Fauna of the brackish-water lake on Verlaten Island* 



The fauna of this small lake of brackish water is of much interest 

 because only recently it became cut off from the sea. The northern part of 

 Verlaten Island is low and sandy land and did not exist shortly after 

 the eruption of 1883. The coast-line was then about where now the hilly 

 land begins (at c, fig. 2). Later on a separate shoal came into existence 

 north of the old coast-line. In May 1Q08 there was a lagoon (at a) still in 

 connection with the ocean; after that this arm of the sea became enclosed, 

 the connection being filled up with sand and now clothed with Casuarine 

 forest; a swampy depression (at b) is all that remains now of the former 

 junction with the sea. The water of this pool is also brackish, but the area 

 is subject to seasonal variations. In October 1919 the ground of pumice- 

 stone was almost dried up and covered with a thin film of salt; the greatest 

 height of the water was observed in April 1920. 



The lake itself (a) is separated from this pool by a narrow and low 

 strip of land, but at hioh water, or with violent winds, the waves of the lake 



