682 William Patten 



facet, or lens, for eaeh ommatidium ; the number of retinophorae iu- 

 creases to four, and two well defined and Constant circles of retinulae 

 are developed, an outer row surrounding the cone, and an inner oue, 

 the stalk of the retinophorae, or the reflector when one is developed. 

 The circles of retinulae may increase, in certain cases, to three or four. 



It will be seen that, as far as the conversion of a so-called double 

 layered ocellus, or stemma, into the compound eye is concerned, I agree 

 more with Lankester than with Grenacher; but there the resemblance 

 ceases; Lankester has, in other respects, foUowed Grenacher in re- 

 garding the crystalline eone cells as homologous with the corneal hy- 

 podermis of the simple eyes. In confirmation of bis supposition, Lan- 

 kester alleges that there is a tendency in the simple eyes to become 

 retinulate; this is especially true in regard to Scorpions, and upon this 

 fact he seems to base bis conclusions, As we bave showu, this 

 statement is true not only of Limulus and Scorpions, but is 

 likewise the case in ali eyes, since ali are formed of omma- 

 tìdia showinga greater or lessamount ofsegregation. The 

 relation of simple eyes to compound is not shown by a 

 greater or less degree of retinulation, since this is carried 

 nearly to its extreme in such remote iormB 2lb Limulus snid 

 Scorpio, but by the position and number of the rods, which 

 tend to become axial, giving rise to the crystalline cone of 

 the compound eye. 



Lankester has introduced a number of terms which we cannot 

 accept. The term merve end cells«, if my observatious are true, 

 is not logicai, since every celi is supplied with nerves ; we can only dis- 

 tinguish the mann er in which they terminate. If we reject the term 

 jmerve end cells«. we must do the same for the term »perineural 

 cellsff. The primary division he would make into »monostichous« 

 and j)diplostichous(f is no more appropriate than the terms simple 

 and compound eyes, or ocelli and faceted eyes. It is, fìrstly, ex- 

 tremely doubtful whether a »monostichous« Arthropod eye exists; and, 

 secondly, many of the so-called »diploblastic« ones are in reality three 

 layered, as in Scorpions (Graber), and Spiders (Grenacher). Ali 

 organs tend to vary in certain directions, dependeut primarily upon the 

 primitive structure of the organs in question. To know what those 

 directions are we must know the structure of a few specialized forms, 

 and how they diifer from the primitive ones. I bave endeavored to 

 find these factors in order to found a new method of Classification. I 

 bave shown, with a certain show of reason, what the originai structure 



I 



