3o8 



NA TURE 



[January 25, 1900 



be the shagreen granule of a remarkable fish — shark-like though 

 not a shark ; for while possessed of an expanded pectoral and 

 an apparently heterocercal caudal fin, of an extensive branchial 

 apparatus consisting {hr. ) of seven or eight sets of parts, it revealed 

 no trace of teeth, jaws, fin-spines, or pelvic fins ; while median 

 fins other than the caudal could not be recognised. The 

 impression created by this announcement was profound, and 

 from the context of his remarks it became evident to the 

 zoologists present that there had been discovered new and 

 probably annectant forms of unique value, and that their best 

 hopes of the Scottish Silurian deposits might yet be realised. 



Shortly after his return to Edinburgh, the lecturer produced 

 the first of the two memoirs under review, and therein an- 

 nounced that the fish in question is that described some thirty 

 years ago by Powrie, its discoverer, from the Lower Old 

 Red Sandstone at Turin Hill, near Forfar, by him named 

 Cephalopterus, and which he himself, noting that that name 

 was preoccupied, had in 1896 re-named Turi/tia. Further, he 

 announced the very important discovery of its generic identity 

 with Agassiz's Thclodus, and, retaining Powrie's specific name, 

 put it finally forward with adequate description and illustration 

 as Thelodtis Paget, a member of the family Coalolepidse, of 

 the nature of whose representatives it enabled us to form the 

 first comprehensive idea. 



This discovery of the generic relationship between the so- 

 called Cephalopterus and Thelodus furnished the keynote to 

 the main series of observations which followed, and the 

 author very properly made the Thelodus Paget 

 the subject of a first special memoir, intro- 

 ductory to the second. Towards the con- 

 clusion of this he lays special stress upon the 

 peculiar characters of that which he regards 

 as its pectoral fin-fold, which, according to 

 his description, would appear (Fig. 1, pc' pc") 

 to extend along the branchial free border, and 

 to be " continuous anteriorly with the outline 

 of the head " ; while, of its posterior lobe, he 

 remarks that the "lappet-like expansion" 

 which it forms "suggests an analogy with 

 the cornual flaps of Cephalaspis, which were 

 originally considered by Lankester as the equi- 

 valents of pectoral fins." To this, to our 

 reading the most noteworthy, passage in the 

 whole monograph we shall return. Passing 

 on, the author, pointing out that while the 

 Coelolepid in most respects resembles the 

 Pteraspidian its dermal covering points to an 

 Elasmobranch affinity, concludes that through 

 the Thelodus allies the Heterostraci may have 

 had a common origin with the primitive Elas- 

 mobranchs — an argument previously formulated by Reis, except 

 that he regards the Pteraspidians as degenerate. 



So much for the first memoir, which, if it alone contained 

 all that has come to hand, would have marked an epoch in our 

 knowledge of these most mysterious of fish forms. 



Turning now to the second memoir, it may be said that the 

 materials described in it were only obtained in 1897, and that 

 the fact that it was not until the author received and compared 

 these with the Powrie specimen that the identity of that became 

 evident. Hence the delay in the final determination of this. The 

 1897 collection was obtained by Messrs. Macconochie and Tait 

 while searching the Silurian Rocks of the Lesmahagow district 

 of Lanarkshire, and it was by the latter gentleman supplemented 

 in 1898 by material which included one of the new genera. 

 The total yield has been five genera, of which four are new, and 

 eight new species, and in the earlier part of the monograph 

 the author, retaining the ordinal name Heterostraci for the 

 Coelofepid and Pteraspidian forms, extends the definition of the 

 former, regarded as a family of the order, so as to include 

 Thelodus and the first of his new genera, Lanarlda{Y'\^. 2). Of 

 Thelodus two new species ( T. Scolictis and T. plautts) are 

 recorded and described, and of Laitarkia three species. All 

 are small, and Lattarkia is characterised by the presence of a 

 dermal armature consisting of relatively large conical spines, 

 without a basal plate {cf. Fig. 2). 



For the reception of the remaining three genera, the author 

 has found it necessary to create a new family, the Birkeniidae, 

 arid a new order, the Anaspida, setting aside for the moment 

 yet another remarkable form, for which the formation of a new 



family of the Cephalaspidians or Osteostraci has been found 

 necessary. To deal firstly with the Anaspida, the most remark- 

 able character which the two genera {Birkenia and Lasanius) 

 possess in common is the possession of a series of median ventral 

 scutes extending along the greater length of the trunk, and bear- 

 ing each, throughout the posterior series in Birkenta (¥\g. 3) 

 and their whole extent in Lasauius{¥\g. 4), a broad compressed 

 spine of formidable aspect. Beyond this very remarkable character 

 they differ in toto. Birkenia, possessed of a heterocercal caudal 

 fin and a single dorsal, which (rf.), with its body, is encased in 

 a dense armature of elongated scutes, at first sight suggests, 

 as the author naively remarks, a Palasoniscid " with the rows of 

 scales running the wrong way ! " The characters of its head 

 scutes, the absence of recognisable mouth, of teeth, jaws and 

 operculum, the non-certainty of orbits, however, altogether out- 

 weigh these superficial appearances as criteria of affinity, and 

 Dr. Traquair, dealing en passant with a series of apertures, which 

 lie along the post-cephalic boundary and which he thinks may 

 be respiratory, with justification makes out a case for an ostraco- 

 deirmatous kinship. Lasanius is still more remarkable ; since, 

 beyond the row of scutes and the heterocercal tail fin, upon which 

 its presumed affinities with Birkenia are based, the specimens 

 by which it is "represented agree only in revealing a series of eight 

 post-cephalic skeletal rods, which (r.) slope obliquely forwards, 

 and by a forwardly directed series of processes approach their 

 fellows of the opposite side near the dorsal middle line. They 

 areserially disposed behind an oblique chain of ossicles (r.'), and 



Lasanius problentnticiis, restored outline, about natural 



remarkable in the extreme as are these structures, there is no 

 suggestion forthcoming as to their probable significance. The 

 author points out that Lasanius stands "much in the same 

 relation to Birkenia as the nearly naked Phanerosteon does to the 

 other genera of Palseoniscidse," and we must be content to accept 

 this on his authority, as justification for his provisional association 

 of these strange genera with the Ostracodermi, pending further 

 discovery, which can alone help us towards a fuller solution of the 

 problem they present. 



The remaining genus described as new is Aieleaspis, for which 

 the author also creates a new family of the Cephalaspidian order 

 Osteostraci. This genus is small, and beset by " shagreen bodies " 

 which on the head have " coalesced into small polygonal plates " 

 that in the tail region unite to form " flat rhombic scales." It is 

 further remarkable for the joint possession of the configuration 

 of a Thelodus or Lanarkia, and of a couple of crescentic 

 markings on the top of its head which the author is inclined to 

 regard as the "outer margins of a pair of orbits, placed as in 

 Cephalaspis " (such as he has failed to detect in any of the other 

 associated genera described in his monograph as new). 



Passing on to general questions of classification and final con- 

 sideration of the inter- relationships of these noteworthy creatures, 

 Dr. Traquair hesitates, to our thinking rightly, to admit the Pter- 

 ichthyidoe members of the Sub-Class Ostracodermi, apparently 

 retaining this for the reception of the Pteraspidian and Cephal- 

 aspidian forms and their allies alone. The main outcome of 

 his work is the association of the former with a supposed 

 Plagiostome ancestry through the Coelolepidae, and he is led to 

 find connecting links in the Drepanaspidse, upon the Pteras- 



NO. 1578, VOL. 61] 



