314 MEiMOlKS OF THE NATIONAL ACADEMY OF SCIENCES. 



He tlien adds : 



The part of the brain formed by the external groove becomes the optic ganglion of the lateral eyes, as in the case 

 of the spider. The optic gaagliou of the median eyes is produced from the epiblastic thickening forming the nnited 

 mouth of the invagination for the median eyes (Fig. 69, g. med. e., our fig.). 



The brain, which was almost circular in its outline, as its height was almost equal to its breadth (Figs. 12, 34) 

 becomes aiterwards semicircular (Fig. 13), later still sickle-shaped (Fig. 14), and shortly before hatching almost 

 straight (Fig. (i9). This is caused probably by the peculiar development of segments. All the segments seem to 

 develop most vigorously at the line midway between the ventral and dorsal median lines, i. «., all the segments have 

 their greatest autero-posterior extension on this line, and become shorter and converge, like the frame pieces of a 

 folding fan, toward both th(! ventral and dorsal median lines. Therefore the external ends of the brain are pushed 

 anteriorly by the growth of the succeeding segments. About two weeks before thehatching of the embryo the brain 

 proper becomes divided into two transverse portions by a constriction (Fig. fi9). , Thus the brain of Limulus may be 

 divided into four parts; the ganglion of the median eyes, the ganglion of the lateral eyes, and the anterior and pos- 

 terior portions of the brain proper. These four parts are arranged in three tran.sverse rows. The anterior row is 

 occu]>ied by the two pairs of the optic ganglia, the middle row by the anterior portion of the brain projter, :ind the 

 posterior row l)y the posterior portion of the brain proper (Fig. 69). 



We have quoted Kisliinouye's observations at length, because they clearly and simply explain 

 as no one had done bdurc, the mode of origin of the ganglia of the brain. In Prof. Kingsley'.s last 

 article little is said upon this subject, his attention not having apparently been directed to the 

 development of the giinglia of the brain. 



My own imperfect observations conlirm those of Kishinouye, but were made on a stage appar- 

 ently before that represented by his Fig. 13, and after that of his Fig. 12. The legs in my embryos 

 had not yet become divided at the ends into claws, while there were three pairs of abdominal 

 appendages. It also appears to be of a later stage than that represented by Kingsley in his Fig. 32, 

 though the ambulatory or cephalothoracic legs of my embryos seem less advanced in development. 

 In my embryos also the neuroblasts had disappeared, though I made repeated attempts to dis- 

 cover them. Kishinouye states that " they disappear when the ventral nerve cord is divided into 

 ganglia and begins to be separated from the eplblast" (p. 73). 



Kingsley's Fig. 33 well represents the cerebral and lateral-eye lobes ; but his sketch does not 

 give a clear view of the median-eye ganglia, which in my specimens were distinctly separated from 

 the lateral-eye ganglia and the cerebral lobes, and were somewhat conical in shape. I reproduce 

 on PI. XVII, some camera sketches of the lobes of the brain seen in the stage represented by the 

 shaded Fig. 1, the month at this time being situated between the first pair of appendages. 

 The eggs were taken from the vicarious chorion or deutovum, anfl stained from two to three minutes 

 in alcohol cochineal, so as to bring out quite clearly the ventral plate and parts of the germ, which 

 had been freshly killed. All the flgiires on this plate were made from a single embryo. 



In Fig. 2 the relation of the lobes is shown by a camera sketch made in direct sunlight. The 

 lateral-eye lobes or ganglia are seen to be distinct from the cerebral ganglia [cer. gang.), and in front 

 of the former are the triangular median- eye lobes. I could not detect any traces of a second pair of 

 median-eye lobes or median eyes; and, as already stated, I am led to think that Patten was mis- 

 taken in supposing that they exist. Jnst above the mouth (w) is a shield-shaped or somewhat tri- 

 angular convex area, which is not figured by Kingsley, and of the nature of which I am doubtful, 

 unless it be the rostrum or upper lip. It was seen very distinctly and at different times. In front 

 of it is a small depression (probably the neuropore of Patten), and a larger one is seen in front of 

 the nnion of the cerebral lobes. 



Fig. 3 is another .sketch made from the same egg, showing the same parts from a side view. 

 In this figure there is represented an irregular depression between the second pair of al)domiual 

 appendages, which, however, was not always seen. Fig. 3ft is a camera sketch of the head of the 

 embryo seen from one side, the left side being the part most distinctly seen. 



The neuromeres were distinctly developed, a pair to each pair of appendages. In the abdomen 

 thiee pairs were distinctly seen, as rexjresented in Figs. 4, 5, and C. I was able to see them very 

 distinctly at different times. In Kingsley's Fig. 32 no abdominal neuromeres are represented. His 

 figure may represent a later stage, though I hardly understand why he does not represent the 

 third pair of abdominal legs, which I repeatedly saw, my Fig. 5 being a careful sketch of those 

 of the right side made in direct sunlight. In Fig. 6 the vent or proctodanim was observed, situ- 

 ated between tlie last two neuromeres. * 



