HARDWICKE'S SCIEN CE-GOSSIP. 



107 



has been the chief agent in excavating the dales,, 

 first by cutting out subterranean channels for itself 

 along the joints of the limestone, and then, the roof 

 of the cavern having falling in, by gradually widen- 

 ing the caverns, and so converting them into dales. 

 Good examples of this kind of action are to be seen 

 at the base of lugleborough, in Yorkshire. 



Sufficient has been thus said to show tliat in the 

 neighbourhood of Castleton there are spots replete 

 with interest both for the palaeontologist and for the 

 physical geologist, since both will find in this beau- 

 tiful portion of our country localities where oppor- 

 tunities are afforded for studying the fauna and the 

 l^hysical structure of the carboniferous system from 

 the millstone, or Kinder Scout grit, down to the 

 lowest beds of the mountain limestone. 



[The student who desires thoroughly to investi- 

 gate the locality touched upon in this paper is 

 strongly recorameuded to consult the Memoir 

 upon the Lower Carboniferous Rocks of the Geo- 

 logical Survey, published by Longmans.] 



Rev. W. H. Paintek. 



THE FORM OF THE DIATOM ERUSTULE. 



THE forms of the Diatom frustule frequently 

 offer difficulties to the young student of these 

 organisms. I am frequently asked to identify forms 

 with which the finder is already acquainted, the 

 reason being that the frustule lies in au unusual 

 position. If the form of the diatom-valve had been 

 rightly understood, this difficulty would not have 

 occurred. 



A frustule consists of two valves and a connect- 

 ing zone or band ; a pill-box with lids affords per- 

 haps the most familiar idea of the diatom cell. The 

 lid or valve of the diatom, unlike the pill-box lid, 

 consists of one piece only, and is more or less 

 convex. The discoid forms offer little difficulty, 

 with the exception of the Campylodisci : in this 

 genus the valves are twice bent in opposite 

 directions, and in consequence of this double flexure 

 all the valves of the Campy lodisci^are necessarily cir- 

 cular. The Naviculaj are also easily understood. 

 The valves of the lanceolate and linear forms are 

 more or less canoe-shaped and more or less deep ; 

 the orbicular being much shallower than the linear 

 or lanceolate forms. 



The most difficult forms to understand are those 

 belonging to the following genera: — Amphora, 

 Cymbella, Amphiprora, Nitzschia, and Surirella. 



The first three forms belong to the Naviculoid 

 group: inNavicula we have the zone connecting the 

 two valves of equal width all round the frustule, a 

 central median line, a central and two terminal 

 nodules ; in Cymbella we find a departure from the 

 true Navicula, the margins of the valves are dis- 

 similar, one side being convex and the other straight 



or concave, the median line and nodules are more 

 or less sub-central, the connecting zone is no longer 

 of an equal width all round, but is narrower on the 

 concave side; in the genus Amphora the median 

 line and nodules are in some cases almost marginal, 

 and the connecting zone so much widened out at 

 the convex margin that both valves of the frustules 

 are visible at the same time. 



The genus Nitzschia is distinguished by the 

 longitudinal keehlike elevation of the surface of the 

 valve. This is, however', never central ; in some 

 forms it is marginal, and in others sub-central ; 

 the connecting zone is not perpendicular to the 

 margins of the valve, but is produced in an oblique 

 direction. 



Fig. 131. Section Fig. 132. Section Fig. 133. Section 

 ot Navicula. of Cymbella. of Amphora. 



In the genus Amphiprora we have a true Na- 

 viculoid type, the valves more or less deep, the 

 median line (or rather keel) central and sigmoid, 

 the connecting zone of equal width all round. 



In Surirella we find a great diversity of outline. In 

 S. intermedia the valve is narrow and sigmoid ; in 

 some of the varieties of S.fasfuosa they are nearly 



Fig. 134. Section 

 of Amphiprura. 



Fig. 135. Section 

 of Nitxschi a. 



Fig. 136, Section 

 of HuaixsUa^ 



orbicular; the connecting zone is of nearly equal 

 width all round the valve (in some of the ovate 

 forms it slightly decreases in width as it approaches 

 the small end of the valve). The principal charac- 

 teristic of the Surirellee is the more or less produced 

 margin forming a frill round the edge of the valve. 

 The above diagrams will perhaps make the 

 foregoing remarks clearer. They represent ideal 

 sections of the frustules of the different genera; 

 the dotted lines represent the connecting zone ; the 

 faint transverse ones the lines of suture or the 

 place where self-division occurs ; and the thick black 

 lines the valves :— Fig. 131, section of Navicula ; 

 fig, 132, Cymbella; fig. 133, Amphora ; fig. 134, 

 Amphiprora; fig. 135, Nitzschia; fig. 136, Surirella. 

 — F, K. 



