82 E. T. NEWTON ON FISHES* TAILS. 



this enables us to approach the subject in a much more satisfactory 

 manner. 



Prof. A. Agassiz traces the development through 12 stages (vide p. 

 83), and according to his observations — 1. A flounder (Pleuronectes) 

 just hatched has the notochord extending into the middle of the tail, 

 which is nearly equal above and below. This is the condition which 

 Prof. A. Agassiz calls the Leptocardial stage. 2. Shortly after the 

 end of the notochord is seen to make a bend upwards, becoming 

 slightly arched below. The very slightest trace of a division is seen 

 in the edge of the fin towards the extremity, and this increases in 

 the next stages. 3. Two plates of tissue are to be seen below the 

 notochord. The division between the embryonic and permanent 

 caudal fins becomes more distinct, and fin rays are to be seen. 4. 

 All the last-mentioned points become more distinct, and the tail is 

 distinctly bilobed. 5 — 9. The permanent tail becomes larger in pro- 

 portion to the embryonic caudal lobe, so that the former soon equals 

 and then surpasses the latter in size, and in stage 9 the em- 

 bryonic caudal is only seen as a little lobe on the top of the 

 permanent caudal. Internally the structures are essentially as in 

 stage 4, but have become more marked. 10. The last stage before 

 ossification ; the embryo caudal is almost gone ; the broad plates of 

 cartilage have assumed more of the form we have been familiar 

 with in the Gasterosteus. In stages 11 and 12 ossification has set 

 in, and the vertebral column has become divided into its component 

 vertebras, and eventually the upturned end of the notochord is en- 

 cased in the urostyle, as in the Gasterosteus. 



The specimen of a young sprat's tail represented in Plate III, fig. 

 1 , includes the last four vertebras of the column with the representa- 

 tives of those which are turned upwards and the hypural appendages ; 

 but in the drawing all the fin-rays are omitted so as to render the re- 

 maining parts more distinct. The figure is enlarged about 25 times. 

 It will be noticed that these four vertebras decrease in size a little, 

 but only a little, as we trace them backwards. They each bear well- 

 developed, neural and haemal arches ; but while the upper arches are 

 about equally developed, the last two lower ones are enlarged and 

 flattened at their extremities. The fifth vertebra is much modified, and 

 commences the upward turn of the vertebral column, which, behind 

 this point, is directed upwards at a considerable angle. Behind the 

 fifth vertebra three or four ossified segments are to be traced. The 

 first of these (sixth from front) seems to form a definite vertebral cen- 



