270 NATURAL SCIENCE [October 



arm-skeleton ; our fossil, however, has an arm-skeleton consisting of 

 a narrow loop, which projects forward into the shell cavity, and this 

 places it in II. B. c. 2. The first sub-division of this is according to 

 the length and curve of the loop, and our brachiopod agrees with the 

 second paragraph, "Loop recurrent, long. 11." Turning to 11, we 

 read " The loop free," which does not agree with our specimen ; and 

 then, " The loop again fixed to the median septum of the small valve. 

 16." Reference to 16 again gives us two sub-divisions, the first of 

 which includes shells that are "Smooth, . . . have large foramen, 

 and rudimentary deltidium. 17." Number 17 includes 3 genera, 

 Kinqena, Magas, and Rhynchora. Kingena has " median septum in 

 the large valve," and a "cross-band connecting the recurrent loop- 

 hands " ; these structures are not found in our fossil. Bhynchora has. 

 " hinge-line straight, long, large valve with area " ; this also does not 

 fit. We are therefore restricted to Magas, and find in fact that the 

 specimen agrees with the characters here ascribed to that genus. 

 Now this is admirable, and as scientific as it is possible for such keys 

 to be. But how often will the student or the field-geologist have a 

 specimen of Magas pumilus showing all, or even a few, of the necessary 

 characters ? Not one specimen in a hundred shows them. In tact 

 Professor Koken himself says of the Brachiopoda : " Since the delicate 

 calcareous bands are usually destroyed or only discoverable _ by 

 laborious preparation, other characters have to be used in practice." 

 In short, give the student a decent work of reference, such as David- 

 son's Monograph or the " Paleontologie Franchise," and he will have 

 determined genus and species long before you have made up your 

 mind whether the specimen has a brachial skeleton at all. The 

 truth is that the principles of classification are one thing and the 

 methods of fossil-spotting are another. The first essential for the 

 latter is an extensive acquaintance with specimens. Any collector 

 of Chalk fossils can tell Magas pumilus if he has once seen it, When 

 he has this acquaintance, then he can proceed to the true knowledge 

 required for the best systematic work. We must learn these concrete 

 sciences like we learn a language : get a good vocabulary first, and 

 proceed to the structure and syntax afterwards. 



The second section of the book gives short diagnoses of the chief 

 species characteristic of the various formations, and is to be used after 

 i »ne has determined the genus. It is inevitably incomplete, and chiefly 

 intended for German students. Even for the fossils of Germany it is 

 not to be relied on without confirmation by the more complete original 

 monographs ; and this being so, it is a pity that there are no references 

 to literature. The fortunate collector of a Taxocrinus rhenanus cer- 

 tainly should not be able to identify it as a Cyathocrimis, the genus in 

 which Dr Koken leaves it. Some of the genera and species to which 

 reference is made, especially among the Gastropoda, we have been 

 unable to discover in literature at all, and have a strong suspicion 

 that they are here introduced for the first time (e.g., Ectomaria, p. 395). 

 This is undoubtedly the case with the name Amorphocystites, intro- 

 duced in a footnote (p. 41 1) as proposed by Jaekel for Caryocyslis 

 testudinarius Von Buch and C. pumilus Eichwald. It is very doubt- 

 ful if any such change of nomenclature be needed ; and in any case this 

 hole-and-corner method of bringing out new names has never yet been 





