136 THE FOKAMINIFERA 



them as true PateUincF, and the author has previously 

 followed up their study on these lines (see ' Bibl. 

 List,' No. 139). The subarenaceous group is never- 

 theless so important and distinct in the Cretaceous 

 and Tertiary rocks that their separation seems to be 

 necessary, especially as we follow the principle of 

 classification by shell structure, which is the most 

 convenient, as we have already seen, for a clear and 

 systematic survey of the order. 



The sub-family of the Loftusiin.e comprises two 

 genera only, in which the chambers are partially 

 occupied with cancellated or labyrinthic developments 

 of the shell-wall. The minute structure of the test 

 in Gyclcmimina represents the same type of structure 

 seen in the shell-wall of the other genus Loftusia. 

 The latter form, however, attains a relatively 

 gigantic size compared with the majority of Forami- 

 nifera ; and although i;t has been regarded somewhat 

 doubtfully as a rhizopod, chiefly on account of its size, 

 it is now generally accepted as a true foraminifer. In 

 both genera the interiors of the chambers are more or 

 less filled up with the cancellated and labyrinthic 

 shell-wall and septa. The microscopic appearance of 

 the central part of CTjclmnmina and the first few 

 whorls of Loftusia are strikingly similar. The con- 

 tour of the test in Loftusia resembles Alveolina, and 

 it may therefore be regarded as an isomorphous form 

 of that genus. 



A third genus was originally included in this 

 sub-family — namely, ParJieria, from the Cambridge 

 Greensand (Cenomanian) — but the structure of its 



