THE RESTORED SHANIDAR I SKULL—STEWART 477 
the abnormal malar portion had to be related to the left temporal 
bone, which in turn had very meager connections with the base and 
occipital bone. As each fragment was placed in position, it was 
anchored by wires cemented to it. When the relationships appeared 
to be correct, the missing parts were filled in with modeling clay. If 
at this stage the eye detected malpositioning, changes were made. 
Ultimately the clay was replaced with “crack filler” and all support- 
ing wires hidden from view. 
This brief account of the procedures used in reconstructing this 
ancient skull has not followed the true sequence of events. Actually, 
the lower jaw was assembled first, then the right side of the neuro- 
cranium (plus the anterior part of the frontal bone) and the right 
side of the face separately. Thereupon the lower jaw was mounted 
in its natural position on a piece of plate glass so that it could be 
turned about easily. The right side of the face was then placed so 
that the teeth of the right side were in natural occlusion. The teeth 
were cemented together (temporarily) in this position. Next the 
right side of the neurocranium (with attached frontal bone) was 
brought into position. In this process attention was given to the 
frontomalar suture, the zygomaticotemporal suture and the temporo- 
mandibular joint. Unfortunately, there was no contact in the naso- 
frontal region, so I had to judge the relationships of the supraorbital 
ridges to the jaws, etc., by eye. After various trials, the frontomalar 
suture appeared to be the connection of prime importance. From here 
on the midplane of the face and neurocranium was clearly established 
and could be maintained in the vertical position. Subsequently, the 
left side of the face, the left side of the neurocranium, the base, ver- 
tex, and occiput were adjusted and added in that order. 
Since by far the most difficult part of the whole operation centered 
about the placement of the left temporal bone, it is desirable to in- 
clude a few words about how this was done. The first step was to 
bring the temporal bone and the left side of the base into position. 
This was a delicate procedure, because this part of the base was dis- 
placed downward. Also, the temporal had lost most of the petrous 
portion, but was nearly complete externally. Thus the mandibular 
condyle had to be relied on to orient the glenoid fossa and make it 
correspond in position with the one on the right. However, in 
addition at least three other parts had to be considered: (1) A piece 
of occipital squama (including the left side of the foramen magnum) 
with a tenuous connection to the temporal; (2) the left side of the 
basiccipital with attached left occipital condyle; and (8) a small 
section of base connecting the temporal and basioccipital anteriorly. 
In fitting all these together the objective was to maintain the position 
of the left temporomandibular joint and to obtain a symmetrical 
foramen magnum. The maneuvering of these fragments would have 
