Manchester Memoirs, Vol. Lx. ( 1 9 1 6), No. 2. 2 1 



of this paper, but to other workers, as no modern restored 

 drawings of an Osteolepid have been published. Our 

 figure represents the most complicated arrangement of 

 bones in the ethmoidal region which we have met with, 

 other specimens showing a reduction in the number of 

 elements obviously due to fusion. These bones, whatever 

 their number, seem to be symmetrically placed. 



In the small specimens from the Thurso Beds in 

 Caithness belonging to Osteolepis microlepidotus, Pander, 

 the whole fronto-ethmoidal region is fused into a solid 

 shield showing only very slight traces of sutures although 

 its shape is similar to that of the older specimens of 

 Osteolepis macrolepidotus described above. 



One specimen on slab No. L. 8425 shows very clearly 

 that the enlarged ridge scales behind the dorsal fins really 

 form part of the general body squamation. 



The fact that it is still unfortunately impossible to 

 recognise any distinct lines of descent even of the 

 broadest kind in Palaeozoic Crossopterygians renders their 

 discussion a matter of considerable difficulty, and its 

 results of an unsatisfactory nature. The figures and 

 descriptions in this paper will serve to substantiate the 

 general belief in the close relationship of all the animals 

 which Smith-Woodward included in his sub-order. 

 Rhipidistia, but they show that along with this general 

 similarity in structure goes a far larger variability in 

 detailed skull structure than has previously been suspected. 

 We have been able to show the existence of several 

 undescribed bones in the heads of Holoptychius and 

 Glyptopomus and the problem now arises whether these 

 fish with a very complex skull have arisen from types like 

 Osteolepis with a simple one or vice versa.. In Tetrapods 

 it is a general rule that evolutionary change has led to a 



