726 General Notes. [July, 
On the other hand the earlier stages in the embryology of 
Limulus resemble those of Crustacea in the absence of the pro- 
cephalic lobes; in the primitive development of cephalic appen- 
dages alone; the comparatively early appearance of the branchiæ 
of Limulus in the stage succeeding that figured in this essay, 
shows that the Limulus enn never had any genetic connec- 
tion with a tracheate arthro 
On the other hand, the pina features of mesoblastic 
somites are also seen in the worms, in Peripatus and in Annelida. 
It appears that the embryology of Limulus is scarcely more 
like that of tracheates than. Crustacea ; it is a very primitive type 
standing nearer the branchiate arthropods than the tracheate, but 
on the whole should be regarded as a generalized or a composite 
form, which with its fossil allies, the Eurypterida and Trilobita, 
form a class by themselves with a superficial resemblance to the 
Arachnida. 
It seems to us that the above-mentioned characters, which sep- 
arate the early embryo of Limulus from the tracheates, are as 
important, if not much more so, than those of the absence at 
first of an archenteric cavity or differences in the mode of origin 
of the mesoblast, noted by Mr. Kingsley in his brief paper on the 
development of Limulus. In these general, primitive embryonic 
characters Limulus appears to be as nearly allied to the annelids 
as to the tracheates; and too much dependence should not, it 
seems to us, = placed upon them in ae to establish the true 
relations of the Paleocarida among the arthropods. In the 
higher worms the two longitudinal mesoblastic bands split into 
somatic and splanchnic layers (K y). In Mysis Metsch- 
nikoff states that the mesoblast becomes broken up into distinct 
somites (Balfour's Embryology, 1, 436). If so, then this charac- 
ter is not one of much importance to separate Limulus from the 
Crustacea. The ultimate origin of Limulus from the same stock 
as that which gave rise to the modern annelids seems not im- 
probable. 
EXPLANATION OF PLATE XXIV. 
Fic. 1 Eae gee Raye (bl. cut) lying Sas the ibang (ef). The nuclei 
scattered throu: e latter; the nucleolus in these as well as the mesoblast cells, 
consisting of a nu oe of granules. x i A. 
Fic, 2.—Longitudi po ory through an embryo before the appearance of the ab- 
dominal append. but after the eee e of the chorion; the section passes 
through the six e k appendages (1 ee showi Sp somatic cavities (7s), 
the iplanclinoplcare (sf), and somatopleu ith 1-9 the indications of the five 
ve uromeres; Ay, hypo- or ecto blast. 
` 
Fic. 2a.—Showing the relations of hypblastic ie Ay) to the epiblast in the 
dorsal region e the emb: : > P soe 
Fic Cees Locse » section of ‘the head and the first three appendages ; ms!, ms?, 
~ firstand second somatic cavities in the preo elipt of the head. This figure 
=: also shows the relations of rae splanchnopleure and somatopleure to the epi- 
large distinct x 4A, 
reps. c, 
Fie. 4.— Aeri ieai ka the head, indleding the e apean X44. 
