THE ANATOMY OF POLYPTEKUS. 331 



fin-rays. The muscle of the anal fin is greatly enlarged in the male, protruding as a 

 rounded mass into the ccelom. It is this mass which causes the ureters and genital 

 ducts to turn so abruptly ventralwards in the male (PL L. figs. 2, 5, 6, an.f.m.). 



I may mention that in an abnormal male specimen in which the anal fin was absent 

 this muscle was completely absent, and the excretory and genital ducts ran backwards 

 into an extension of the ccelom, then forwards ventrally to open in the normal position 

 behind the vent. 



This sexual character almost entirely disappears out of the breeding-season. 



Leydig has suggested that there is internal fertilization in Polypterus from the fact 

 that in the ccelom of a female Polypterus he found masses of filaments which he took 

 for spermatozoa. That these filaments are not spermatozoa can at once be seen by 

 comparing the figure he has given with my figure of spermatozoa from the ripe testis 

 (PI. LI. fig. 13). 



When the arrangement of the oviducts is considered, it seems extremely improbable 



that the spermatozoa would find their way into a duct which opens into the urino- 



genital sinus upon a papilla. It remains to be seen to what use the male Polypterus puts 



this modified anal fin. 



IX. The Vascular System. 



The blood-supply to the external gill has been worked out by Hyrtl, while the main 

 roots of the arterial system were described by Joh. Mailer. I would call attention, 

 however, to a few additional details. Having injected a male specimen with salt- 

 solution when killed, it was re-injected with a coloured gelatine in the laboratory. 

 The specimen was adult and had no external gill. The details of the blood-supply 

 to the external gill were made out on a young specimen in which it was possible to 

 inject the hyoidean artery with a coloured fluid. The figure of the arterial system was 

 made by a combination of these two dissections. 



As Hyrtl has shown, the hyoidean artery arises at the anterior end of the ventral 

 aorta immediately in front of the first afferent branchial artery, and passes to the 

 base of the operculum, at the centre of which it meets the efferent hyoidean artery, 

 to run with the latter to the posterior edge of the operculum and thence to the 

 external gill. 



The point I wish to call attention to is that the afferent and efferent arteries at 

 the extremity of the gill are continuous one with another, forming a drawn-out 

 loop. From the afferent limb branches run to the pinnee, at the extremity of 

 which they loop back to the main efferent limb ; similar tertiary loops pass into the 

 pinnules. 



At the root of the external gill there is a dorsal and a ventral muscle ; towards the 

 extremity of the gill these break up into numerous isolated bundles (PI. LII. figs. 

 25, 26, 27). The whole arrangement is quite similar to that of an Amphibian or 

 Dipnoan external gill. 



vol. xv. — part vii. No. 2. — April, 1901. 3 a 



