EUROPEAN PORITES. 107 



wall, or else first as a septal granule or plate, the septa having a tendency to be lamellate. 

 Here and there fusings of septa take place, but quite irregularly. The pali are also very 

 irregular, appearing only as stray radiating points or plates, with a small, slightly flattened, 

 columellar tubercle. 



In the exposed cross-sections of this coral there is a solid layer of nearly 3 mm. thick 

 surrounding a very thick strand of stout, lamellate, streaming layer. The solid cortical 

 layer seems to have been due to subsequent mineralogical changes, because a close examina- 

 tion of it shows stout, rather regular trabecules and concentric elements here and there still 

 visible. 



There are indications, further, that, when alive, the living layer extended only a few 

 centimetres from the tips. 



This Porites has all the ordinary habits of a West Indian branching form, and, as further 

 Mr. E. A. Smith kindly informs me that the molluscs * indicate the same locality, I have no 

 hesitation in placing this specimen in this group. 



a. Geol. Dept. E. 9820. 



Group VIII.— EUROPEAN FOSSIL FORMS. 



On turning eastwards from the North Atlantic Porites of Bermuda, we look in vain for 

 any recent forms on its eastern shores. The only records are of rich fossil remains of a Poritid 

 fauna which inhabited the Paris and Mediterranean Basins in early and mid-Tertiary times, and 

 the Vienna Basin and the Crimea as early as the deposition of the lower Cretaceous strata. Very 

 few of these, however, appear to have been true Porites, although the majority were at first 

 assigned to that genus. The records given in detail in Vol. IV. show how great was the con- 

 fusion between Porites and Goniopora ( = Litharcea). The majority of the forms belong to the 

 latter genus rather than to the former, that is, if the lines of division here adopted prove to be 

 correct. 



One of the chief difficulties in the way of the scientific classification of these fossils, lies 

 in the fact that their .structural details, which in this genus are especially intricate, are so 

 often obscured. In many cases we have no chance of ever seeing a surface preserved well 

 enough to show the walls and septal formula of the calicle. In most cases we only have 

 sections, and from them, in the normal process of growth, all these surface characters upon 

 which classification depends disappear. It is therefore obvious that no classificatory method, 

 dependent on exact morphological details, such as the method by " specific" or " form " units, 

 is at all possible, for the necessary data are not available, being either absent or blurred by 

 change. On the other hand, let me say that nothing is so dangerously facile as the tossing of 

 these fossils into so many imaginary species, and the extent to which that has been done in 

 the past, renders much of the work hitherto published absolutely valueless. One such wholly 

 imaginary "species," "P. incrustans" we shall have to notice in the brief list which follows; 

 we shall call attention, not only to its purely imaginary character, but also to its disastrous 

 effect upon our knowledge of the forms it was thought to help to classify. If some more exact 



• See note (f) on page 106. 



P 2 



