52 INTRODUCTION. 



All naturalists are aware that linear arrangements do not occur in nature, therefore 

 the present as well as all similar attempts must be viewed as artificial, and more or less 

 conventional ; our endeavours have, however, tended to group together those animals 

 which in the present state of our knowledge seem to possess essential similarity of character, 

 and have, therefore, provisionally divided the class into ten families and four sub-families ; 

 but it must be observed that when Palaeontologists are better acquainted with the internal 

 characters of the genera and species, the limits, as well as the number of the groups, will 

 most probably require modification, and the more so since those forms located for the 

 present in sub-families have not been sufficiently worked out to warrant their admission 

 in the groups already defined. No small amount of investigation has taken place in the 

 endeavour to find out what principle or what characters should be considered of sufficient 

 importance to authorise the separation of families and genera ; the difficulty will be easily 

 understood when it is remembered that out of forty sections or subsections here introduced, 

 fourteen alone possess living representatives, and the animal of several of these last has 

 not hitherto been completely anatomically examined. The difficulties, therefore, naturally 

 increase in those cases, unfortunately by far the most numerous, where all that can be 

 gathered of the inhabitant of the shell consists in a few impressions existing in the 

 interior of the valves. The study of the animal of Terebratula, Crania, Discina, Lingula, 

 and a few other recent types, has to a certain extent opened the way to a knowledge of 

 portions of the animal of many extinct genera, and it is by analogy and careful com- 

 parisons that modern Palaeontologists have perceived the absolute necessity of separating 

 certain things which early naturalists had confounded together, from not being able to 

 interpret or understand the use or importance of the calcified processes or impressions 

 visible in the interior of the shell. 



All Brachiopoda are possessed of two valves articulated by the means of teeth and 

 sockets, or iinarticulated, and kept together entirely by muscular action. Out of the forty- 

 five sections or sub-sections enumerated, thirty-three are distinctly articulated, but in one 

 of the most natural groups, viz. that of the Productidce, we find genera presenting both 

 conditions. The animal of the Brachiopoda lived entirely free, or was fixed to marine 

 l)ottoms during a part or the whole of its existence either by the substance of a portion of 

 the ventral valve, or by the means of a muscular peduncle passing between the valves, or 

 through an aperture existing in the ventral or dental valve, it is probable that the majority 

 of the species were fixed, at least in an early stage of their existence, however much they 

 may have dispensed with that temporary attachment at a more advanced period. In the 

 same family and even genus, however, we discover forms which had been attached, while 

 others present no traces of such a condition. It would, therefore, be impossible to separate 

 the attached and unattached species, genera, and families, from this character alone. The 

 sha[)e and position of \}a& foramen, or its absence, when combined with other peculiarities, 

 may help to define and circumscribe certain groups. The value to be attached to the shell 

 structure has already been ably discussed by Dr. Carpenter, that we need only remind our 



