BY E. HAVILAXD. 71 



Considering that only about five per cent, of the flowers open, 

 and consequently that that proportion of the pollen only is avail- 

 able for the fertilization of the plant, and that the anthers do not 

 open by slits, as is most usual ; but by two very minute pores, 

 thus giving out the pollen grudgingly ; that the pollen, even in 

 the anther, is by no means abundant, but rather the reverse; 

 taking, too, into consideration that each flower produces but one 

 seed, one cannot help feeling surprise that the plant is so 

 abundant. I think, however, this may be accounted for to a great 

 extent, by the presence of the secretion of which [ have spoken. 

 The fact of its being more copious in the younger flowers while 

 the styles are comparatively short, and the secreting glands are 

 near the surface of the head of flowers, renders it almost impossible 

 for an insect to crawl about the stage, formed by the compact head, 

 without becoming so smeared with it, that on visiting an open 

 flower much of whatever pollen might be exposed, would adhere 

 to it, and be carried away by it ; and for the same reason upon 

 its visiting the mature stigmas of the closed flowers, a little only 

 of the pollen, but cpiite sufficient, since there is only one ovule to 

 be fertilised, would be left on each ; the greater part remaining 

 adhering to the insect, who would carry it from one flower to 

 another. Thus it is not unlikely that by the pollen obtained by 

 one visit to an open flower, very many of the closed ones would 

 be fertilised. 



Studies ox the Elasjiobranch Skeleton. 



By "William A. Haswell, M.A., B.Sc. 



[Plates I. and II.] 



In his well-known memoirs " Das Kopfskelet der Selachier " 



and "Die Brnst-Flossen der Fische,"f Gegenbaur has treated 



very exhaustively of the structure and homologies of these part s 



* Untersuchungen zur vergleichenden Anatomie der Wirbelthiere. Heft- 3. 

 t Op. cit. Heft. 2. 



