42 ON THE SPECIES OF FAVOSITES OF THE 



have a diametei' of one and a half lines, whilst some few have a 

 diameter of a line or a little less. 



Whilst the mural poi-es are usually in two rows, there is sometimes 

 but a single row, and sometimes three rows ; and single colonies may 

 be found to combine all these variations in different corallites. It is 

 probable that the typical forms tipon which Goldfuss founded his 

 species F. hasaltica, as believed by Lonsdale and McCoy, are truly 

 referable to sj)ecimens of F. Gothlandica, Lam., in which but a single 

 row of pores is present. F. Golclfussi, Edw. & H., again, seems 

 unquestionably to be nothing more than a variety of F. Gothlandica, 

 as asserted by Mr. Billings, the only distinctive characters brought 

 forward by its authors being the unreliable ones that the rows of 

 pores vary from one to three, and that they are more closely approxi- 

 mated than in typical examples of the latter. Lastly, the elevated 

 ring which is often fotmd surrounding the pores in F. Gothlandica, 

 is in many cases absent, single specimens often exhibiting both 

 conditions. It may be added that the pores themselves, even in 

 specimens otherwise well preserved, can by no means universally be 

 detected. 



The tabulae are for the most part complete, extending from one 

 side of the theca to the other ; but this condition of parts is by no 

 means constant. Specimens, otherwise well preserved, sometimes 

 exhibit a complete absence of the tabulae, the corallites being hollow. 

 Others exhibit a condition of things very similar to what occurs in 

 Favosites Forbesi, though not so marked. The inner surfaces, namely, 

 of the corallites exhibit rudimentary and imperfect tabulae, in the 

 form of little projecting lamellae, or ridges, which project into the 

 cavity of the theca. Specimens exhibiting this peculiarity can 

 usually be distinguished from examples of F. Forhesi without diffi- 

 culty, hy the fact that the ridges representing the tabulae are not so 

 closely set, are more delicate and plate-like, usually run across the 

 whole width of the corallite, and do not give to the interior of the 

 theca the extraordinary roughness of appearance which is character- 

 istic of F. Forhesi. In other specimens, again, the tabulae have the 

 characters which are distinctive of F. hemis'pJierica, being closely set 

 and incomplete, often more or less bent, and commonly interlocking. 

 Such specimens, however, are readily separated from those which are 

 i-ightly referred to F. heniispherica, by the fact that in the former 

 some of the corallites are always found to exhibit the complete tabulae 



