34 BRITISH STROMATOPOROIDS. 



component parts of the skeleton, and thus yield results which may be easily 

 misleading. 



Moreover, under certain circumstances, which are not easy to explain, the 

 skeleton of the Stromatoporoids is liable to undergo a more or less complete 

 secondary crystallisation, by which the internal structure is greatly obscured 

 or, it may be, completely obliterated. This is seen in many of the specimens from 

 the Devonian Limestones of Devonshire, and particularly in many of those found 

 in the rolled limestone pebbles of the Triassic conglomerates of South Devon. In 

 these cases, it would seem probable that the crystallisation is largely connected with 

 mechanical causes, as it is almost always accompanied with a greater or smaller 

 amount of distortion of the skeletal framework. In other cases, however, the 

 crystallisation is the result of an internal rearrangement of the particles of which 

 the skeleton is composed, the general form of the skeleton remaining unchanged, 

 while the surface and the epitheca may be beautifully preserved. This is commonly 

 seen in the Stromatoporoids of the Wenlock Limestone of Gotland and of Esthonia, 

 and, more rarely, in specimens from the Wenlock Limestone of Britain. 



In the following general account of the structure of the skeleton of the Stromato- 

 poroids, it will not be possible to take any one single type as illustrative of the 

 main facts to be considered, as there exist very wide variations within the limits of 

 the group, as here understood. We shall find, however, that these variations may, 

 on the whole, be reduced to one or other of two leading types of structure. In 

 one series of forms, of which the true Stromatopora concentrica of Goldfuss is the 

 type, the skeleton is of -what maybe called the "Milleporoid" type, having what Mr. 

 Carter has designated as a "curvilinear" structure. In the other great series of 

 forms, typified by Actino stroma clathratum, Nich., 1 the skeleton is of what may be 

 termed the " Hydractinioid " type, having what Mr. Carter has called the 

 " rectilinear " structure. 



The bond of union by which these two groups of forms are linked together, is 

 found in the fact that the calcareous coenosteum in both groups can be shown to 

 be made up of two sets of elements, one vertical to the surface, and the other 

 parallel with the surface. In the " Milleporoid " series, typified by Stromatopora, 

 Goldf., the vertical or " radial " elements are so combined with the horizontal or 

 " concentric " elements as to give rise to a continuously reticulated skeleton, in 

 which the elementary constituents are with difficulty recognisable as distinct 

 structures. On the other hand, in the " Hydractinioid" series, typified by Actino- 

 stroma, Nich., the "radial" and "concentric" elements of the skeleton remain 

 more or less clearly recognisable as distinct structures, and the skeleton never has 

 the form of a continuous vermiculate reticulation. 



1 It is to be borne in mind, as previously explained, that the form here called Actinostroma 

 clathratum is what has hitherto been regarded as being Stromatapora concentrica, Goldf., a microscopic 

 examination of the original of the latter having shown that its structure is quite uulike what it was 

 supposed to be. 



