156 Geological Society : — 



rounded form, and consists of a few closely packed but well-deiined 

 cells, which are very much shorter than the other cells of the hair. 

 The elongated cells of the body of the hair (of which the lower one 

 is most characteristic on account of its very large size), contain fine 

 colourless, granular matter, and generally nuclei ; but the secreting 

 cells are well furnished with colouring matter of a reddish-brown, 

 but sometim.es of a green colour. A one-inch object- gla^s, recom- 

 mended for the examination of tobacco, is usually insufficient to 

 show the structiwe of the gland, and the mere presence of * glandular 

 hairs' proves nothing, these being common in plants. It is also 

 necessary to keep in view that many small hairs occur on tobacco- 

 leaves which are normally without glands. The glandular hairs are 

 most abundant at the tips of the shoots, and especially on the calyx 

 and flower-stalks of the tobacco. To the fact that epidermal hairs 

 are so frequently organs of secretion, Gasparrini has recently added 

 the additional one, that they are also the organs of absorption. 



■ 4. "Notice of Galls found on the Leaves of the Beech/' by Mr. 

 James Hardy. 



GEOLOGICAL SOCIETY, 



January 6, 1858. — Major-General Portlock, LL.D., President, 

 in the Chair. 



The following communications were read : — 



1. "On Cephalaspis and Pteraspis," By Prof. Huxley, F.R.S., 

 F.G.S. 



Of the four species originally included by Prof. Agassiz in the genus 

 Cephalaspis, two, C Lloydii and C. Lewisii, are so different from the 

 others that the possibility of their proving generically distinct is 

 hinted at in the ' Recherches sur les Poissons Fossiles.' 



Subsequently M. Kner endeavoured to prove that these two species 

 are not fish-remains at all, but are the internal shells of a Cephalo- 

 pod, for which he proposed the generic name of Pteraspis. 



Roemer has still more recently expressed the opinion that the 

 Pteraspides are Crustacea. These conflicting opinions clearly in- 

 dicate the necessity of revising and comparing anew the characters 

 of the diff'erent species of Cephalaspis and Pteraspis. And a still 

 greater interest is lent to the inquiry into the true nature of Pte- 

 raspis, from the fact that species of this genus are now known to 

 occur in undoubtedly Upper Silurian rocks. As the evidence stands 

 at present, they are, if fish, among the oldest (and nearly the very 

 oldest) representatives of their class. 



In undertaking this inquiry, the author of the present paper con- 

 sidered it desirable, in the first place, to determine with precision the 

 microscopical characters of the shield of Cephalaspis. This shield is 

 exceedingly thin, nowhere exceeding -j-Vth of an inch in thickness on 

 the dorsal surface, and on the ventral suddenly thinning off* a little 

 way from the margin into a mere membrane. 



The subjacent cranium appears to have been wholly composed of 

 cartilaginous or soft fibrous tissue ; for the *' layer of fibrous bone," 



