LIMULUS AN ARACHNID, 639 
Scorpion, and there is no heart which so closely resembles 
the Scorpion’s as does that of the King Crab. 
That the internal reproductive organs should open ex- 
ternally in the neighbourhood of limbs is certainly not a 
peculiarity of Crustacea. The relation of the openings to 
limbs is not ‘much as in Crustacea,’ but quite unlike any- 
thing seen in Crustacea. In no Crustacean does a pair 
of limbs in front of the genital apertures unite to form 
with a median lobe carrying those apertures—a broad plate, 
as in the King Crab. A genital operculum of this nature 
is found only in the King Crab, the Kurypterina, and the 
Scorpion. 
The extreme anterior position of the generative apertures 
has no parallel among Crustacea nor among Arthropods, 
excepting the Arachnida, where it is identical in position. 
Even the chilognathous Myriapods do not exhibit so forward 
a position of the genital orifices. 
K. Conciusion ; LimuLus AND THE ANCESTRY OF 
TRACHEATE ARTHROPODA. 
The nature and degree of intimacy of the relationship 
between Limulus and the Scorpion—which is indicated by 
the facts and arguments set forth in the preceding essay— 
have yet to be considered. It is one thing to establish the 
fact that a closer relationship obtains between Limulus and 
Scorpio than between Limulus and any Crustacean, and 
another thing to estimate more precisely the affinity between 
the two animals. : 
A brief consideration of the facts is sufficient to show that 
the points in which Limulus agrees with Scorpio and Mygale 
include those structural features on which we have to rely 
in attempting to characterise the class Arachnida. At the 
same time it must be admitted that all attempts at limiting 
classificatory groups by simple definition are hopeless, pro- 
vided that the groups are intended to express degrees of 
genealogical affinity, and not merely arbitrary categories, 
held together by more or less obvious class marks. The real 
question which we have to attempt to answer, in assigning 
Limulus and the Arachnida their place in a genealogical clas- 
sification of the Arthropoda is not, “ How may groups be de- 
fined which shall give due expression to the structural 
likenesses and unlikenesses of these forms ?” but, “‘ How may 
groups be arranged so as to exhibit the probable history of 
ancestral development in relation to these forms?” Owing 
to the occurrence of degeneration, and to the suppression in 
