110 DEVONIAN PAUNA. 



inferior corner. Surface covered with numerous sharpish, rather irregular 

 transverse lines and undulations of growth. 



Size. — Specimens measure 55 mm. long by 55 mm. high by 35 mm. deep ; or 

 40 mm. long by 35 high by 20 deep. 



Localities. — Numerous specimens are in the Woodwardian (including Sowerby's 

 types) and the British Museums, the Barnstaple Athengeum, the Museum of 

 Practical Geology (including the seven specimens figured by Phillips^), and Mr. 

 Hamling's and my own collections from Sloly, Marwood, Shirwell, Baggy Point 

 and Knowle Quarry near Braunton, North Moulton (one), and Tiverton (four). 



Remarks. — Sowerby and Phillips have divided the Cucullsese, which are so 

 common in, and so characteristic of, a zone in the Marwood beds, into five or six 

 species ; but it may be observed that at least Professor Phillips had great doubts 

 as to the correctness of so doing. 



Of G. angusta, C. unilateralism and G. trapezium he writes, " The three last 

 species are so very closely allied, that on looking over many specimens we find it 

 doubtful whether the differences observed are other than those of degree. The 

 same thing is observable at Marwood in regard to Avicvla, damnoniensis, which 

 varies so much in its proportions as to require the pointing out of three distinct 

 forms." 



M'Coy, uniting G. amygdalina to G. nnilateralis, and G. trapezium, to G. Eardingii, 

 reduced Phillips's five species to three. But G. amygdalina is intermediate 

 between G. angusta and G. unilateralism and G. Hardingii bears much the same 

 relation to G. angusta which G. amygdalina bears to C. trapezium. It may be said 

 indeed, on the other hand, that these five forms, though bearing evident relation- 

 ship, are easily distinguishable, and can be definitely defined, and that it is possible to 

 class specimens under them. But, even so, it soon appears, when any large number 

 of specimens are under examination, (1) that hardly any two that may be placed 

 under any of these forms are exactly alike, showing that each group is merely a 

 collection of very variable shapes bearing some resemblance in general features ; 

 and (2) that many specimens may with equal justice be classed under more than 

 one of the five forms : that is to say, that they not only vary within themselves, 

 but run in all directions into each other. It seems, therefore, necessary to believe 

 that we have really only a single very variable species, and that the five forms are 

 not even local varieties, but simply accidental variations of shape. It must further 

 be remembered that these fossils have been subjected to a good deal of squeezing and 

 pressure, and that consequently some amount of the variability of the specimens 

 may be due to that, and therefore have no real zoological existence at all. Lastly 

 it may be remarked that the five forms described by no means exhaust the shapes 

 which the species assumes ; and that the only consistent course is either to 

 > 1841, Phillips, ' Pal. Toss.,' pp. 40, 41, pis. xviii and xix, figs. 66—70. 



