Biology of the Social Wasps of Brazil. 137 



The Old-World genera not mentioned in this list all belong 

 to the first group. Divers species of Polyhia are known from 

 tropical Asia, but I know nothing about their life-history. 



In conclusion, I would add a remark as to the genital 

 apparatus, in which the closest agreement exists in all the 

 genera of social wasps that I have investigated. The 

 number of testicular tubes is always the same as that of 

 those of the ovary, and this number is invariably three on 

 each side, as is also the case in the solitary wasps examined. 



In the strongest contrast to this condition, that I have 

 proved to exist in Polistes, Mischocyttarus, Nectarinia^ 

 Polyhia, and Chartergus, is what we have recently learnt 

 from Bordas *, in amends for previous inexact statements. 

 According to this author the number of testicular tubes in 

 Vespa is from two hundred to three hundred. How large the 

 number of the ovarian tubes really is (earlier accounts speak 

 of six or more) must remain uncertain in default of fresh 

 examination. Vespa consequently presents'a strong contrast 

 to the rest of the social wasps, and in this respect stands in 

 the same relation to them as that in which the Apid^e stand 

 to the Andrenidge or Megachilid^e, since in the former there 

 are three hundred testicular tubes, in the latter three. It 

 appears that for the Hyraenoptera this number three repre- 

 sents the primitive condition, and therefore Apis and Vespa 

 are aberrant types. As to the anatomy of the social wasps 

 of Africa and India we as yet know nothing. In these cases 

 the work accomplished for the South-American genera has 

 still to be done. Then, and not until then, will it be possible 

 to ascertain the relation of Belonogaster, Icaria, &c. to the 

 rest of the Vespida. Probably it will then be found that 

 Vespa cannot be included in the same family as the other 

 genera of social wasps. The true Vespidte, including Vespa, 

 would then have a very large number of testicular tubes ; 

 the rest of the Polybiid^e would exhibit three of these struc- 

 tures in each testis. It is probable that upon closer investi- 

 gation yet other differences will be added in other organs or 

 in the larvae, and to point out the necessity for such studies 

 is precisely the object of this exposition. 



Sao Paulo, Brazil, 

 July 23, 1896. 



* Bordas, " Appareil pjenital male dea Hym^noptfires," Annales des 

 Sciences natur. vii. 66r. torn, xx., 1895. 



