CHONDROCRAXIUM OF THE LARVAL HERRING. 1215 



this centre; corrections to allow for this can quite easily be 

 made from longitudinal sections of a similar larva. To make 

 comparison more easy and more accurate, all recoiistructions 

 were made to the same scale, using the same micrometer and 

 microscope. 



3. External Characters. 



The specimens examined varied in length from 5 to over 

 50 mm, ; the 10 mm. stage was taken in the greatest detail, and. 

 it corresponds to an age of less than a month. The larvpe were 

 of the Herring, distinguished from those of the Pilchard and Sprat, 

 as is pointed out by Fage (1920), by the number of myotomes, 

 having 56 to 58, 51 to 52, and 46 to 48 respectively. 



The tail is characteristically short, the anus being very close to 

 the end of the body. In a specimen 10*5 mm. long we have the 

 following measurements : — head 1"5 mm., trunk 7'5 mm., tail 

 1*5 mm. (Lebour, 1921). The head is nearly as broad as it is 

 long (including the eyes), the lower jaw protruding fomvards well 

 beyond the upper. 



The only visible fins at first are the pectoral and caudal, the 

 pelvic fin not appearing until a length of over 20 mm. is reached, 

 and the dorsal fin is not complete until the animal is 28 to 

 29 mm. long (Lebour, 1921). 



None of the specimens of 10 mm. upwards examined had the 

 yolk-sac still present, although it may persist until 12 mm. 



These and other points have been illustrated in text-fig. 1 . 



Metamorphosis takes place at about 40 mm. 



4. Discussion. 



This research, as stated in the Introduction, was carried out to 

 fill a very noticeable gap in the work done on the development 

 of the Teleostean chondrocranium. The paper does not pretend 

 to be a complete account, and it has very little detail in it ; the 

 results obtained were so unexpected and abnormal that I have 

 made no attempt to explain or to theorise. Time was unfortu- 

 nately too short for me to make it a comparative account ; it is, 

 in fact, merely a simple description of the state of aSairs 

 found. 



The first striking point is this : there is in the early stages no 

 upper jaw at all — it does not appear until the larva attains a 

 length of about 13 mm. 



Another point is that at the earliest stage examined (i. e. 

 5 mm.) the hyomandibular mass of cartilage is fused both to the 

 cranium and to Meckel's cartilage. It breaks away from the first 

 at about 15 mm., and the second at 25 mm. The upper jaw 

 grows back, and fuses with it secondarily. The quadrate is 

 formed from this mass by a constriction forming in it. 



