298 .V. WATASK 



the rhabdoiii the jiroxiiiiul ciuls of tlie vitrclliie apparoiitly 

 teriiiinnto.' 



When the spindlo is iiiiU'oratcd and each of its coiiipoiient 

 elements is examined individually, it is found that oach retinula 

 develops on its median snrlace a tiiiek chitinons border which 

 shows a wavy, corrugated appearance. This chitinons border 

 attains its greatest thickness in the midst of the retinula cell (Fig. 

 35, PI. XXXI). When the chitin-secreting edges of a number 

 of these retinula cells meet they form a spindle-shaped bundle, 

 the thickest part of the rhabdomere corresponding to the most 

 bulging part of the spindle. The lower extremities of the 

 retinula cells send out nerve processes, each of which forms an 

 optic nerve fibre, {Op.n, Fig. 4, PI. XXXI). Fine elongated 

 cells filled with yellow pigment granules are found between the 

 adjacent retinula bundles {ep, Fig. 4). They are also the modi- 

 fied ectodermal cells. The pigment cells which occur around 

 the vitrellae form a complete envelope around the dioptric 

 portion of each ommatidium. 



Thus even with this complicated type of ommatidium the 

 plan of structure is fundamentally alike to that of Serolis. The 

 central element, the " hyaline cells" of Serolis, is not represented 

 in the crayfish. To this special point I shall recur. 



If we plot out the omniatidial elements of the crayfish as we 

 did those of Serolis, exactly a similar kind of diagram will be 

 formed. Thus in the centre will come the group of retinulae, 



' Parker found in Homarus {A Preliminary Account of the Development and 

 Bistolugy of the Eyes in the Lobster. Proceedings ot the American Academy, 

 Vol. XXIV, pp. 34-25, 1888) that the lower extremities of vitrellae run over 

 the spindle, and that each terminates on the retinal side of the basement mem- 

 brane. It is probable that the same is true in Cambanis. This fact is perfectly 

 intelligible according to the view which is maintained in the present paper, for 

 the cells found in the higher level ot a given ommatidium are morphologically 

 situated in the outside of those found in the lower. Patten's account (£yes 

 of Molluscs and Arthropods, Mittheilungen aus der Zoologischen Station zu 

 Neapel, Bd. VI, Heft 4, 1886) that the rhabdom is the continuation ot the 

 retinophorae (vitrellae) is not true. I macerated the "spindle" of Penaeus, 

 upon which Patten bases his observations, and found that it is composed of a 

 number of pigmented retinula cells, each with its chitinons border as in 

 Cambarus (Fig. 3.5, PI. XXXI) or in Ilomarwt ( Pig. 34, PI. XXXI). Grenacher 

 is right in regarding the rhabdom as the secretion product of the retinula 

 cells. 



