ROBERTSON FLOWERS .t INSECTS, ASCL. TO SCROPH. 577 



were found attached to the hairs of the face, labrum, tongue, 

 or ventral surface. 



Hymenoptera. — Api'da', (i) Apis mellifica L. 5,v.s.; (2) Bombus 

 separatus Cr. -^ ^ [^ , ab., 1., t.. v.s. ; (3) B. scutellaris Cr. ^ , ab., v.s. ; 

 (4) Megachile brevis Say 9 ! (5) ^^- mendica Cr. $, v.s. Ves/u'dce. (6) 

 Polistes pallipes Lep. Eumetitdw., (7) Odynerus arvensis Sauss. Pki- 

 la7itkida\ (8) Cerceris bicornuta Guer.. f., t.. v.s.: (9) C. compar Cr. 

 Bembecidie, (10) Bembex nubillipennis Cr.. ab.. 1. Sphecidiv, (11) Prio- 

 nonyx atrata Lep., v.s. Pompilidn', (12) Priocnemis unifasciatus Say, t., 

 v.s.; Scoliida\ (13) Myzine sexcincta F. 



Coleoptera. — Scarabceidiv, (14) Trichius piger F.. v.s. 



Lepidoptera. — R/iopalocera, (15) Thecla sp. ; (i6) Chrysophanus thoe 

 Bd.-Lec. Pyralidii\ (17) Scepsis fulvicollis HUbn. 



GENTIANACE^. 



Gciitiana Andreivsii Griseb.* — The flowers are pi'oteran- 

 drous. They remain closed, so that only the largest and most 

 intelligent bees can open them. The stamens are united with the 

 corolla tube, and their free ends bend over upon the pistil. To 

 reach the nectar the bee's tongue must be thrust between the fila- 

 ments and must be 15 or 16 mm. long. The flower is visited 

 abundantly by Bombus amcricaiioruin F. cJ* ? S • 



There is no doubt that flowers were originally of such a form 

 that almost any insects could enter them and i^each the nectar. 

 Many have narrowed the entrance by the development of tufts of 

 hair, or of pi"ocesses like the palates of personate flowers, until all 

 insects were excluded except bees. These have kept on visiting 

 the flowers until now they enter completely closed flowers like 

 those oi Linaria. 



They seem to have been trained to bad habits in this way, for 

 they sometimes force their way into flowers which are not yet 

 ready to receive them. I have mentioned the case of Bombus 

 vagans forcing her way into the closed buds of Triostcum 

 perfoliatum^ and B. americanorum does the same in the case 

 of buds oi Li)iaria vulgaris. Mr. Pammel refers to a number 

 of cases in which flower-buds have been opened or perforated by 

 insects.! The irregular behavior in such cases is not to be laid 



* See Beal, Am. Nat. viii. i8o, and Vausenburg, ibid. ix. 310. 



t On the Perforation of Flowers, Trans. St. L. Acad. Sci. v. 2SS> note. 



V. — 3 — 16 



