188 BULLETIN: MUSEUM OF COMPAKATIVE ZOOLOGY. 



Somewhat later Eeissner's fibre develops a thin medullary sheath of 

 myelin throughout its length. As there is no trace of either Heule's or 

 Schwann's sheath, the myelin must necessarily be a secretion of the axis- 

 cylinders. If any medullary sheath is developed about the separate 

 axons where they are free, it is so sliglit that it has not yet been 

 distinguished. 



It is pi'obable that the axons of the tectal reflex cells running caudad 

 have entered the cord before Reissner's fibre has become completely con- 

 solidated and developed its medullary sheath. Numerous appearances 

 in ray sections lead me to think so. But it is not until the eighth day 

 in Amia that I have seen what I could positively identify as axons enter- 

 ing the cord. Tliey then come off as delicate branches of Reissner's 

 fibre, and pass through the canal obliquely (usually making an angle of 

 about 45° with the fibre) and enter the cord from the ventral portion of 

 the canal. They can frequently be traced passing between the ependy- 

 mal cells which line the walls of the canal, and for a short distance into 

 the cord, where they ai"e lost. I believe there is no doubt of their con- 

 nection with the ventral root and musculature, evidence of which of 

 another kind is given elsewhere in this paper. I liave no knowledge, 

 however, as to whether they pass out of the ventral root directly to the 

 musculature, or whether they connect with another neuron. The 

 branches from Reissner's fibre are supposedly the axons of the tectal 

 cells, thougli as to their course througli Reissner's fibre, I only know that 

 it shows a fibrillar structure. 



h. Adult. From the 30-day larva, 2 cm. long, to the adult, I have 

 studied no intermediate stages. Altliough there is a great develop- 

 ment in the mesencephalon as a whole, tlie optic reflex apparatus re- 

 mains in statu quo, except as the development of the optic lobes aff'ects 

 its relations. Comparing the adult Amia (Fig. D) with the 30-day 

 larva (Fig. C), we see that the tectum has increased greatly in thick- 

 ness, owing to the development of the cellular elements ; but near the 

 median plane it is due more particularly to the growth of the optic tracts. 

 The tectum has grown so as to bulge cephalad and dorsad from the pos- 

 terior commissure, consequently increasing the size of the mesocoelic 

 recess in the median plane above the posterior commissure. At the same 

 time the posterior commissure has developed posteriorly between the 

 optic reflex cells and the ependymal thickening (Figs. C, D), so that the 

 ependyraal thickening has apparently receded cephalad. As a result of 

 these morphological changes, the optic reflex cells have come to lie on 

 the anterior and lateral walls of the recess. It is into this recess that 



