CHARLES W. METZ. 



89 



fifty, representing in both cases many widely separated locali- 

 ties, and including many large series. Only with such series 

 as this can a fair idea of any species be obtained. 



With the preceding facts in mind we can more readily 

 understand why the older descriptions of Prosopis species are 

 so unreliable and confusing. The work of the oldest writers, 

 Smith, Kirby, and others, is practically useless, except as 

 supplemented by later descriptions. Following the natural 

 tendency they characterized their species by the general 

 form and amount of the yellow markings and the general 

 size of the insect, these being the most conspicuous features. 

 The characters which they were accustomed to use in other 

 genera were lacking in this, and they consequently took the 

 most conspicuous they could find. The bees being practi- 

 cally naked no use could be made of pubescence, and being 

 so homogeneous in the structure of the legs no distinctions 

 could be based on these. Consequently they overemphasized 

 the color characters and size. The latter is such a variable 



TRANS. AM. ENT. SOC. , XXXVII. (12) 



