NOVTTATES ZOOLOGICAE XXVT. 1919. 21 



Strand, 1 with 3 spots = tripunctata Strand, and lastly one with 5 spots, but 

 not one with i spots. The naming of aberrations on such sUght ground appears 

 to me to be useless, especially as from the large series in the Adams collection, 

 those at Tring and many other collections, the specimens with 1, 2, 3 or more 

 spots appear to occur in almost equal numbers. The ab. comhijmta is more 

 distinct owing to the reduced median white band on hind wing. 



3 <J(J, 2 ?? Brazil (1 ex coll. Felder) ; 1 ? South America ; 1 ? Lapa, Brazil 

 (5 submarginal spots) \ i SS Province Rio ; 1 <? Rio de Janeiro ; 1 <? Interior 

 of Brazil (ex Vienna Mus. ex coll. Felder) ; 1 (J, 1 $ ab. combinata Strand, 

 (J Brazil, ? Parte Allegro, Brazil ; 1 $ Rio Grande do Sul. 



Castnia (Prometheus) garbei Foett. 



Castnia garbei Foetterle, Rev. Mus. Paul. vol. v. p. 639. pi. xvi. fig. 6 (1902) (Rio Grande do Sul). 



Messrs. Houlbert and Oberthiir had never seen this species, but Mr. Talbot 

 records (see antea) Q SS and 3 ?? in the Joicey collection and also some in the 

 Adams collection. 



We have 5 at Tring : 



3 (JcJ, 2 ?? Rio Grande do Sul. 



Castnia (Prometheus) houlberti sp. nov. 



It is most extraordinary that 3 such closely allied species as cochriis, garbei, 

 and houlberti should occur in Southern Brazil. 



Differs from both cochrus and garbei in the forewmg being deep black ; a 

 broad obUque transverse band slightly paler, more sooty and less densely scaled, 

 down the centre of which runs an intense black line. In cochrus this line is inside 

 the pale band and in garbei outside the pale band. A complete row of white 

 submarginal spots on hindwing as in garbei and the central band as in cochrus 

 ab. combinata. Eggs like grains of wheat. 



1 ? Rio de Janeiro (Arp coll.). 



Castnia (Orthia) therapon Koll. 



Castnia therapon KoUar, Ann. Wien. Mus. vol. i. p. 218. pi. xiii. fig. 3 (1839) (Rio de Janeiro). 



1 ^ Rio de Janeiro ; 1 <? 1 (ex Berlin Mus. ex coll. Felder). 

 Castnia (Cyanostola) diva Butl. 



Castnia diva Butler, Lepid. Exot. p. 46. pi. xvii. ff. 1, 2 (1869—1874) (Chontales). 



Messrs. Houlbert and Oberthiir rightly separate tricolor Feld. from diva 

 Butl., but I think no one else can possibly consider it more than a snbspecies, 

 while they accord it full specific rank. Messrs. Houlbert and Oberthiir have 

 taken Dr. Strand's maculifera to be the race from Chiriqui, while they say they 

 do not know his Chiriquensis. Dr. Strand's figure of maculifera shows the red 

 margins broken into spots like Monsieur Oberthiir's Chiriqui o, but 1 have 2 (J (J 

 from Chiriqui with the band entire though suffused with dark scales. It is there- 

 fore evident that specimens with broken red bands occur in both Central 

 American races, while we can only guess that it may be the same in d. tricolor, 

 for only 1 (J is as yet on record of that race. 



The Tring Museum has 9 diva diva : 



1 <J, 1 $ Carillo, Costa Rica, May— July 1903 (Underwood) ; 1 ? Guatemala 

 (Salvin, ex coU. Felder) ; 2 o" J ? ; 4 <J<J, 1 Carillo, 3 Esperanza, May (W. Schaus). 



