68 JRemews — Dr. Traquair on Silurian Fishes. 



flat, rhombic, sculptured scales. The orbits seem to have been 

 close together on the top of the head, as in Cephalaspis. 



A new order, Anaspida, is established to contain the two last 

 and perhaps most extraordinary new genera treated in the memoir. 

 These comprise small, laterally compressed organisms, without 

 paired fins, but with a single small dorsal fin and a well- developed 

 heterocercal tail. In many respects they are suggestive of Cepha- 

 laspis and its allies, but there is no continuous head-shield, and 

 the orbits are not clearly distinguishable in the known specimens. 

 The substance of the armour is not sufficiently well preserved to 

 furnish satisfactory sections for microscopical examination. Birhenia, 

 with a single species, B. elegans, is the first genus placed here. It 

 is completely armoured with small scutes arranged in the strange 

 manner indicated in Dr. Traquair's careful restoration (Fig. 2). On 

 the side apparently of the back of the head there is an oblique series 

 of eight small round holes, which look like branchial openings. 

 The flank-scales of the trunk are disposed chiefly in series which 

 incline forwards and downwards, not backwards and downwards 

 as is usual in fishes. A median series of ventral scutes develops 

 into a formidable armour of spines in the caudal region. The 

 second new genus, Lasanius, with two species, L. problematicus 

 (Fig. 3) and L. armatus, is considered to be closely related to 

 Biricenia, differing in the loss of all the dermal armour except 

 a few anterior flank-scales and the median ventral row of spines. 



The detailed descriptions of these organisms are illustrated by 

 a beautiful series of five plates, most of the drawings being by 

 Mr. James Green, but the highly magnified details and critical 

 points by the author himself. 



In the concluding section of the memoir, which is a general 

 discussion of the results. Dr. Traquair gives for comparison a brief 

 description of the hitherto problematical Lower Devonian genus 

 Drepanaspis. Its main features are illustrated in an outline-sketch 

 which he kindly permits to be reproduced here (Fig. 4). In shape 

 it resembles Thelodus and Lanarkia, but it is covered with well- 

 developed plates and scales, and the orbits seem to have been placed 

 laterally. All the known plates of Drepanaspis are too much 

 pyritized for microscopical examination ; but Dr. Traquair has 

 shown that the other Devonian plates named Psammosteus belong 

 to a elosely similar animal, and these contain no bone-cells. The 

 Drepanaspidae and Psammosteidee are therefore placed by him in 

 the Heterostraci. 



It follows from these researches that the definition of the order 

 Heterostraci must now be slightly modified. We must admit that 

 the armour may consist either of separate granules (each formed 

 round a distinct papilla), or these granules may be more or less 

 fused together in every degree from small polygonal plates to great 

 shields. According to Dr. Traquair the order will therefore include 

 the known families of CoelolepidEe, Psammosteidee, Drepanaspidae, 

 and Pteraspidae, each exhibiting a progressive modification over the 

 other in the development of the dermal armour. 



