520 PROFESSOR GREGORY ON 



time, I have found a considerable number of new species, exhibiting the same pe- 

 culiarities of structure. 



The general character of all these species is, that the frustule is very convex, 

 and formed, in the first place, of two segments or valves, in shape like Cymbellae 

 or the halves of simple Amphorae, placed opposite each other, as in the simple 

 species. The middle space between these is, in the next place, covered, to a greater 

 or less extent, by a complex mass, like that seen in Amphiprora complexa, 

 (fig. 61), and exhibiting, when the two lateral segments lie flat, a series of con- 

 vergent longitudinal bars, which are the backs of other segments, grouped like 

 those of an orange or a melon. The lateral segments exhibit all the characters 

 of the halves of a simple Amphora ; they have the inner curve lines and the no- 

 dules, in some cases elongated so as to form a cross bar. The segments forming 

 the central mass appear to have neither curve lines, nor nodules, nor bars. The 

 lateral segments are transversely striated, and the bars of the central mass, or 

 backs of the central segments, are frequently also marked by transverse striae ; but 

 in other cases have much coarser markings, approaching more to the nature of 

 puncta or granules. In some species again, while the margins of the lateral seg- 

 ments are very coarsely marked, the backs of the middle segments are much more 

 finely striated. In general, when the frustule lies so, that the two lateral seg- 

 ments are flat or nearly so, we see in one focus the lateral segments with their curve 

 lines and nodules, while in another focus these become invisible, and the conver- 

 gent bars alone are seen, which in many cases fill up the whole space between the 

 lateral segments, of which only a part, that nearest the outer margin, can then 

 be seen. This is shown in several of the figures. 



This structure is so peculiar, that it seems as if it would be desirable to estab- 

 lish a new genus for the reception of these forms. But the form of the seg- 

 ments is so exactly that of the valves (or frustules) of simple Amphorae, and 

 even the entire complex forms, in a certain focus, are so like the others, that I 

 think it will be sufficient to make a sub-genus for the complex forms. 



I have already mentioned that some of these forms occur in the Glenshira 

 Sand ; others, as well as these, are found in the dredgings from Lamlash Bay. 

 But by far the larger part of the complex species here described are to be found 

 in that one Loch Fine dredging spoken of under A. turgida, as being rich in Am- 

 phorae. Of the above list, thirteen species at least, along with A. costata, besides 

 fourteen or fifteen of the simple group, occur in this one dredging, and several 

 of these are found in it alone. It will be unnecessary to repeat this in detail for 

 each species ; and I shall briefly refer to Loch Fine as the locality of the species 

 here alluded to as occurring in that particular dredging. 



82. Amphora lyrata, n. sp. PL XIIL, fig. 82. Form doubly lyrate, with 

 truncate ends, and a notch in the middle, on each side, where the two lyrate 

 halves meet base to base. The lateral segments have each a strong bar or elon- 



