1907.] Wheeler, The Polymorphism of Ants. 23 



to tlie genus Cremastogaster (C. minutissima Mayr and C victima missouri- 

 ensis Pergande). Both are small honey- yellow species which nest under 

 stones in shady places. From a colony of minutissima taken June 3, 1901, 

 in the very same locality at New Braunfels as the parasitized colonies of Ph. 

 commutata, I took three large workers which were described and figured by 

 Miss Holliday as ergatoid females.^ These measured 4.3-5 mm. whereas 

 the normal workers of this species do not exceed 2 mm. They had small 

 ocelli and greatly enlarged gasters, and though no parasites were found in 

 them, it is not improbable that they contained young specimens of Mermis 

 that may have been overlooked when Miss Holliday made her dissections 

 for the purpose of ascertaining the condition of the ovaries. 



In a lot of C. victima missouriensis collected by Miss Augusta Rucker at 

 Paris, Texas, I have recently found four individuals resembling the mi- 

 nutissima mermithergates described by Miss Holliday. They measured 

 ■3.2 and 4.5 mm. respectively, although the normal workers measure only 

 2-2.5 mm. None of these mermithergates has ocelli. In the two largest 

 specimens the gaster is very voluminous but seems to contain no parasites, 

 and the mesonotum is unusually large and convex, so that the thorax resem- 

 bles that of a pseudogynic Formica. 



Recourse to the literature shows that Emery had seen mermithergates 

 in certain Central American and Soutli American ants as early as 1890. 

 In his paper on a collection of these insects from Costa Rica he describes 

 and figures two odd specimens of Pheidole absurda Forel from Alajuela.^ 

 They measured 7.25-7.75 mm., although the normal worker of this species 

 measures only 2.75-3 mm. They had huge, elongated gasters, vestiges of 

 ocelli, and a thorax similar to that of the soldier, but in the shape of the head 

 they resembled the worker caste. He did not observe that they were par- 

 asitized but regarded them as "parthenogenetic females." In the same 

 paper, and from the same locality, he also described three specimens of 

 Odontomachus hoematodes with a huge gaster and a minute ocellus, as possibly 

 a new variety (var. ? microcephalus) .^ In this connection he mentions three 

 other similar cases: an 0. chelifer (var.? lepfocephahis Emery) from Rio 

 Grande do Sul, a Neoponera villosa inversa from Venezuela, and another 

 tropical American ant, Ectatomma tuberculatum. In all of these cases the 

 great volume of the gaster was naturally attributed to an unusual develop- 

 ment of the ovaries. 



Since reading my paper Emery has reexamined his specimens and finds 



1 A Study of Some Ergatogynic Ants. Zool. Jahrb. Abth. f. Svst., XIX, 4, 1903, pp. 311, 

 312, 16 figs. 



2 Stndii .sulle Formiche della Fauna Neotropica. Boll. Soc. Ent. Ital. Ann., XXII, 1890. 

 p. 49, pi. V. figs. 10 and 11. 



3 Emery calls attention to the fact that this form of hccmatodcs had been previously seen by 

 Roger. 



