
No. 425.] THE TREMATASPID.E. 385 
their consequent inversion took place much later, probably in 
the intermediate type of animals to which the Tremataspidze 
belong. It was during this period that the median eyes 
reached their highest development and the lateral eyes degen- 
erated or, in some cases, disappeared completely from the sur- 
face. We have no means of knowing whether their temporary 
decline in functional importance was the cause, or the result, 
of their transformation into eyes of the cerebral type. 
The postorbital opening probably contained the forerunner of 
the vertebrate olfactory organ. I have identified it with the 
frontal organ of Limulus, Branchipus, Apus, and others. This 
organ in the arthropods presents extraordinary variation in its 
position, but can always be identified by the peculiar histological 
structure of the terminal organ and its nerve, and by the origin 
of the nerve in the brain. In Limulus, no doubt, the organ was 
originally a visual organ serially homologous with the lateral 
and median eyes. In the adult it lies on the ventral side in 
front of the chelicerz. It gradually loses the histological 
characters of a visual organ and finally presents many points of 
resemblance to the olfactory organ of a vertebrate, especially 
in the structure and relations of its nerves. The hypostomeal 
. eyes of trilobites are very probably homologous with the olfactory 
organs of Limulus. In Branchipus the same organ has moved 
from its original position on the ventral surface to a point on 
the dorsal surface almost as far back as the median ocellus. In 
Apus the two organs have moved still farther back and have 
united behind the median ocellus to form an unpaired organ in 
precisely the same location as the post-orbital opening of 
Tremataspis. 
The Oral Plates have been worked out anew from the single 
fossil and its cast that was used by Rohon. My description 
differs from his in several very important respects. 
According to my interpretation of this important fossil there 
are on each side nine large oral plates arranged in four rows. 
Some of the marginal plates are provided with one or more 
rounded incisions lying opposite corresponding incisions of the 
ventral and dorsal shields. The marginal plate of the fourth 
row is probably a compound plate. There is a row of four or 
