40f> bulletin: IIUBEUM 01 COMPARATIVI zoology. 



Ml li.Mi < OPHILOl B I'l.w rs, 



Bulbous epiphytes o! the genera Myrmecodia and Hydnophytum, 

 1 found none in Sower bo cannot differentiate between the genera, 

 ait- common In Fiji, They are known to the natives by the name 



"seki-seki," and arc used in the witchcraft ceremony of "drau rri kau." 



One bakes from his enemy a hair from his head, some food Left from his meal 



01 BOme other thing intimately connected with him. This is placed with 



leaves of Myrmecodia and parts of other plant- in a bamboo tube, which is 

 then scaled. A curse is wished on the enemy and as the contents of the tube 



ferment or decompose, he becomes afflicted with a malady, (boils BOem to be 

 the favorite curse) and can only be cured by unsealing the jar and letting tin- 

 scape. There is no doubt that, to the native mind, this is taken seriously 

 and the psychological and resultant physical affects on one who knows he is 

 being bewitched by "drau ni kau," are real. The government has framed 

 laws against the practice. On our walk across \ 'hi Levu, the people in one 

 village gravely warned my boys of the danger in the next village and cautioned 

 them against leaving crumbs of food about. Portions of the bulb are con- 

 sidered a remedy for deafness by the natives about Nadarivat u. 



At the edge of the moss forest by the Taviuni Lake, I found, on the 

 stunted trees, a number of large bristly bulbs. Each of these was 

 tenented by a populous colony of Iridomyrrnex nagasau subsp. alticola. 

 In the forest on the McKenzie estate at Nagasau, another species of 

 bulb, smaller and less bristly was abundant high in the trees. Of a 

 dozen examined, all contained colonies of Iridomyrrnex nagasau. A 

 species at Nadarivatu occurred in similar situations, but, instead of 

 always sheltering one certain species of ant, was used as a nesting site 

 by various species of Iridomyrrnex, Camponotus, Poecilomyrma, and 

 Pheidole, and many contained no ants at all, but myriopods, spiders, 

 scorpions, or geckos and their eggs. 



In the preparation of this paper, Prof. W. M. Wheeler has aided me 

 repeatedly and his collection has been always available for compari- 

 sons. 



The majority of the drawings w r ere made by Dr. R. S. McEwen. 



1. Cerapachys (Cerapachys) vitiensis, sp. nov. 



Worker. Length 3 mm. 



Head a third longer than broad, as broad in front as behind, with moderately 

 convex sides, angulate occipital corners and broadly and shallowly concave 



