MUSEUM OF COMPAKATIVE ZOOLOGY. 25 



with the junction of two retinulee. As a matter of fact, three lines do 

 come between retinulte, but the fourth one abuts against the middle of 

 the large odd retinula (Fig. 34). This relation of the cross-lines and 

 retinula) persists from the earliest stage in the production of the rhab- 

 dome, and although the lines are doubtless formed during this process 

 they show a strange independence of the retinulse. 



An exactly similar relation between the cross-lines and retinulte has 

 been described by Grenacher ('79, p. 124) in Paltemon. I cannot 

 agree with Grenacher in believing that the four lines are due to the 

 fact that only four retinula^ are concerned in the production of the 

 rhabdome. The relations of the retinulse and rhabdome are the same in 

 the young lobster (Fig. 58) in which the rhabdome is being produced 

 as in the adult. This fact was not known to Grenacher. It shows, I 

 believe, that the rhabdome is the product of the surrounding seven reti- 

 nul^e, and that the problematic lines have some other significance than 

 that of indicating regions of production. 



The Accessory Pigment-cells. 



These cells occupy the open space at the base of the ommatidia. 

 They are characterized by possessing a pigment which, as I have before 

 stated, is brownish by transmitted and whitish by reflected light. The 

 cells are bounded pro.\imally by the basement membrane, and their 

 distal ends rarely reach beyond the middle of the rhabdomes (Fig. 1). 

 They are extremely irregular in form, and seem to fit themselves to a 

 cavity of almost any shape. Their function seems to be that of filling 

 what would be otherwise an unoccupied space, as though to lend solid- 

 ity to the tissue in the base of the retina. (Compare Figs. 13 and 16.) 

 Their nuclei are irregular in form and size (Fig. 18, nl. pig.). Judging 

 from the number of nuclei, two or three cells are associated with each 

 ommatidium. The number, liowever, is variable. The physical prop- 

 erties of the pigment which these cells contain are very characteristic. 

 Streaks of this pigment, and even whole cells, are to be met with in 

 the open space on the proximal side of the basement membrane. 



Cells similar to those which I have called accessory pigment-cells have 

 been described by Carrii're ('85, p. 1G9) in Astacus, and by Patten ('86, 

 p. 636) in Pena'us. It is highly probable that the yellow pigment 

 which Grenacher ('79, Fig. 114) figured in the base of the retina of 

 Mysis represents accessory pigment-cells as well as the dark pigment 

 which he described ('79, p. 124, Fig. 117) in Paltemon. 



