SEPTEMBER 25, 1902] 
NATURE 
329 
Passing to the third advance, extending over the last twenty 
years, it may be said to consist in the revival of a theory of 
1829, which boldly asserts that Limulus is an Arachnid. In the 
development of the defence there have been two weak points 
but lately strengthened, viz., the insufficient consideration of 
the palzontological side of the question and of the presence of 
trachee among the Arachnida. Under the former there was, 
until recently, assumed the absence of the first pair of appendages 
in the Eurypterida; but it may be said that they have since 
been observed in Zurypterus Fischeri of the Russian Silurian, 
and £&. scoticus from the Pentland Hills, in both of which they 
consist of small chelate appendages flexed and limuloid in detail, 
somewhat reduced perhaps, and enclosed by the bases of the 
succeeding limbs, which become apposed as the anterior end is 
reached. Since by this discovery the Limuloids, Eurypterids 
and Scorpionids are brought into a numerical harmony of limb- 
bearing parts, we may at once proceed to other points at issue. 
So far as the broader structural plan of Limulus and the 
Scorpion are concerned, all will agree to a general community, 
except for the organs of respiration; but concerning the ccelom, 
the mobile spermatozoa and the more detailed features under 
which Limulus is held to differ from the Crustacea and to 
resemble the Arachnida, I would remark that while motile 
spermatozoa are characteristic of the Cirripedes, the rest of the 
argument is weakened, by the probability that the ‘‘arachnidan”’ 
characters which remain may well have been possessed by the 
crustacean ancestors, and that Limulus, though specialised, 
being still an ancient form, might have retained them. The 
difficulty does not seem to me to lie in this, or with the 
excretory organs, if we are justified in accepting the aforemen- 
tioned argument that the so-called Malpighian tubes may be 
inturned nephridia, ectodermal in origin, and in knowledge of 
the existence of endodermal excretory diverticula in the 
Amphipods. These facts would seem to suggest that as our 
experience widens, differences of this kind will disappear. 
As to the tracheal system, now adequately recognised by the 
upholders of the arachnid theory, the presumed origin of tracheze 
from lung-books, the probability that the ram’s-horn organ of 
the Chernetidze may be tracheal, the presence of trachez in a 
simple form in the Acari, and, by way of an anomaly, in a 
highly organised form on the tibize of the walking legs of the 
harvestmen (Phalangidze), are all features to be borne in mind. 
While I am prepared to admit that this wide structural range 
and varied distribution of the trachez lessens their importance as 
a criterion of affinity, I cannot accept as conclusive the evidence 
for the assumed homology between lung-books and gills. And 
here it may be remarked that a series of paired abdominal 
vesicles, recently found in the remarkable arachnid Kcenenia, 
invaginate as a rule but in one example everted, seized upon in 
defence of this homology, have not been so regarded by those 
most competent to judge. 
There remains the entosternite, an organ upon which much 
emphasis has been placed. Not only does a similar organ exist, 
apart from an endophragmal system, in Apus, Cyclops, some 
Ostracods and Decapods, but, regarding the question of its 
histology, it may be pointed out that, from all that is at present 
known, the structural differences between these several ento- 
sternites do not exceed those between the cartilages of the Sepia 
body. And when it is found that the figures and descriptions 
of the entosternite of Mygale (‘‘ Mygale sp.,” ‘‘ Mygalomor- 
phous Spider,” act.) have been thrice presented upside down ! 
the reliability of this portion of the argument is lessened, to say 
the least. 
Recent observation has sought to clench the homology of the 
four posterior pairs of limbs of the King Crab and Scorpion, by 
appeal to a furrow on the fourth segment in the former, believed 
‘o denote an original division into two; but I hesitate to accept 
this until myological proof has been sought. 
Returning, amidst so much that is problematic, to the sure 
ground of palzontology, I wish to point out that when all is 
considered in favour of the arachnid theory there still remains 
another way of interpreting the facts. 
In both Limulus and the Scorpion the first six of the eighteen 
segments are well known to be fused into a prosoma bearing the 
limbs, but while in the Scorpion the remaining twelve are free, 
in Limulus they are united into a compact opisthosomal mass. 
In dealing with the living arthropods, there is no character 
determinative of position in the scale of this or that series more 
trustworthy than the antero-posterior fusion of segments. It 
has been called the process of ‘‘cephalisation,” and the degree of 
NO. 1717, VOL. 66] 
its backward extension furnishes the most trustworthy standard of 
highness or lowness in a given assemblage of forms. In passing 
from the lower to the higher Crustacea, we find this fusion 
increasing as we ascend : and it therefore becomes necessary to 
compare the Scorpion with the other Arachnida, Limulus with 
the Eurypterida, in order the better to determine the position 
of each in its respective series, by the application of this rule. 
As to the number of segments present, variation is a matter of 
small concern, in consideration of the mode of origin of seg- 
mentation and the wide numerical range—from seven in the 
Ostracods to more than sixty in Apus—the segments of the 
crustacean class present. 
On the arachnidan side, in the Solifugze but the third and fourth 
segments are fused ; the remaining four of the prosomal series with 
the ten which remain are free, In Koenenia four of the prosomal 
segments alone unite ; the fifth and sixth with the rest are free. 
And when we pass to the Limuloids and the descending series 
of their allies, we find it distinctive of the Eurypterida that alb 
the opisthosomal segments are free. If we can trust these com- 
parisons, we must conclude that the Eurypterida of the past, in 
respect to their segmentation, simplify the Limuloid type, on 
lines similar to that on which the Solifugee and Kcenenia 
simplify the Higher Arachnid and Scorpionid type, and that 
therefore, if the degree of antero-posterior fusion of segments 
has the significance attached to it, Limulus and Scorpio must 
each stand at the summit of its respective series. If this be 
admitted, it has next to be asked if, in comparing them, we may 
not be comparing culminating types, which might well be 
isomorphic. 
The scorpions are known fossil by two genera, Palzeophonus 
and Proscorpius, from the Silurian of Gotland and Lanarkshire, 
the Pentland Hills and New York State ; while recent research, 
in the discovery of the genus Strabops, has traced the Eury- 
pterida back to the Cambrian, leaving the scorpions far behind. 
One striking feature of the limbs of the Palzeozoic Eurypterids 
is their constantly recurring shortness and uniformly segmented 
character, long known in Slimonia, and less conspicuously in 
Pterygotus itself, retained with development of spines in three 
of five known appendages of the recently described eurypterid 
giant Stylonurus. The minimum length yet observed for these 
appendages is that of the Silurian species Zurypterus Fischert, 
discovered by Holm in Russia in 1898. This creature is one of 
the few eurypterids in which all the appendages are preserved, 
and it is the more strange therefore that the advocates of the 
arachnid theory should ignore it in their most recent account. 
Allowing for the specialisation of its sixth prosomal appendage 
for swimming, the fifth is but little elongated, the second, third 
and fourth are each in total length less, by far, than the trans- 
verse diameter of the prosoma, and uniformly segmented, giving 
the appearance of short antenne. They seem to be seven- 
jointed, and are just such appendages as exist in the simpler 
crustacean and tracheate forms ; and in the fact that their struc- 
tural simplicity is correlated with the independence of the whole 
series of opisthosomal segments they lend support to the 
argument for isomorphism. 
With this conclusion, we turn once more to the Scorpions, if 
perchance something akin to it may not be in them forthcoming. 
The Silurian genus Palzophonus, especially as represented by the 
Gotland specimen, reveals the one character desired. Its body 
does not appear to be in any marked degree simpler than that of 
the living forms ; but on turning to its limbs, we find the four 
posterior pairs, in length much shorter than those of any living 
species, all but uniformly segmented. In this they approximate 
towards the condition of the limbs of the Eurypterida just dis- 
missed, and their condition is such that had they been found 
fossil in the isolated state they would have been described as the 
limbs of a Myriapod, and not of a scorpion at all. Indeed, their 
very details are what is required, since in the possession of a 
single terminal claw they differ from the limbs of the recent 
scorpions as do those of the Chilopoda from the hexapods. 
With this the scorpionid type is carried back, with a struc- 
tural simplification indicative of a parallelism with the other 
arthropod groups; and while the facts do not prove the 
total independence of the scorpionid and limuloid series, they 
bring the latter into closer harmony with the Eurypterida of the 
past. They prove that the Silurian Scorpions simplify the 
existing Scorpionid type, on precisely the lines on which the 
Eurypterida simplify che Limuloid ; and they do so in a manner 
which suggests that a distinction between the Crzstacea vera and 
the Crustacea gtgantostraca (to include the Eurypterida and 
