SONNINIA— REVISION. 379 



Sonninin^j (continued). 



Genus — Sonninia (continued) . 



Revision of tee Sonninim of the Concavum-zoxE, and Descriptions of New 



Species. 



It has become necessary to change the method adopted in grouping these 

 species of Sonninia. The various investigations necessary for specific determina- 

 tion have revealed that extreme septal differences were hidden beneath great 

 external similarity of form — a similarity which, standing alone, had been considered 

 too great to admit of specific distinction in view of protests, ofttimes unreasoning, 

 against new species. Want of opportunity for thorough investigation is frequently 

 the cause of such protests ; and, in the present case, a certain want of thorough- 

 ness, partly induced perhaps by the remembrance of such protests, led to 

 mistakes as serious as the following of the opposite course would have done. 

 Too many forms were grouped under one specific name ; and it is the object 

 of the following observations on a revised classification to correct the mistakes 

 which have arisen from this cause. 



It was, perhaps, a rash assumption that external similarity of form should 

 be coupled with internal septal likeness. In the case of Dumortieria and 

 Grammoceras a difference of septation accompanies a certain outward similarity ; 

 yet, in most of the species, there are slight external differences which the eye soon 

 becomes expert enough to detect. Such is, again, the case between Sonninia and 

 Hammatoceras : there a difference in the trend of the ribs accompanies a 

 noticeable difference in septation. In both of these cases, then, peculiar features 

 of outward ornament always excite the suspicion of septal differences : in the 

 present instance there was nothing to do this. By chance, however, it was 

 discovered that examples of Sonninia, externally similar in dimensions and 

 ornament, and without any difference in the trend of the ribbing, possessed 

 markedly different septa. Then it became necessary to examine with great care 

 the septation of every specimen. 1 Had this been done in the first place, certain 



1 This proved to be an immense labour. First, the specimens of Sonninia from Bradford Abbas 

 either do not show the septal margins because the test is present, or they show them very badly, because 

 when the test is absent the surface is eroded. When the test is present it is necessary to remove it. 

 When the core is hard and smooth, the test, if it be crystalline and not too securely attached to the 

 core, may be removed by chipping with a small chisel (a fine-tempered bradawl or a strong carpet- 

 needle ground to a chisel-edge about 1 mm. broad) held nearly at right angles to the specimen. If, 

 however, the test be firmly adherent, it can be induced to leave only by repeated small blows with an 



