234 
points {d, e) to a folliculinid now known to 
undergo only binary fission. It is equally absurd 
to assume that the present body attachment 
point (e) permits a better peristome exit. In 
light of possibility I (Neck Extensions by the 
Stage that Secreted the Lorica), two attachment 
points and two neck extensions might be ex- 
plained as follows: the builder of the original 
lorica, once the neck (c) was completed, con- 
tracted into its sac but, after a period of rest, 
instead of metamorphosing into a swimmer, 
relaxed and, while in the lobed stage, secreted 
the first imperfect neck extension ( b ). It then 
withdrew and metamorphosed into a swimmer 
which vacated the lorica. The present body at- 
tachment point (e) is that of a new swimmer 
which entered, became attached, and, while in 
the motile stage, secreted the second extension 
{a). This may explain in part why spiral whorls 
are absent in the first extension (b) and why 
132 jj 
Fig. 5. Metafolliculina andreivsi as viewed from 
the right side showing: a, second neck extension; b, 
first neck extension; c, neck; d, original body attach- 
ment point; and e, present body attachment point. 
PACIFIC SCIENCE, Vol. XVIII, April 1964 
they are present in the second (a). This might 
also account for the fact that the diameter of 
the second extension is approximately half that 
of the first. While these possibilities are not 
conclusive, other examples suggest that neck 
extensions may have multiple origins. 
Frequently M. Nordgardi (Fig. 6A) is ob- 
served in which the body attachment point (d) is 
far removed from the base of the lorica (e). In 
such instances either the body has freed itself 
from its original attachment point ( e ) and 
become reattached (d), or another swimmer 
has entered the old lorica and established itself. 
Because many M. nordgardi with and without 
extensions (Fig. 6B-C) possess loricae whose 
lengths exceed that illustrated in Figure 6A and 
yet experience no difficulty in extending their 
peristomal lobes, it seems rather unlikely that 
reattachment in the shorter form was the result 
of necessity. Moreover, if d (Fig. 6A) was the 
original body attachment point there is no way, 
based on our present knowledge, to explain the 
formation of that portion of the lorica between 
d and e. 
If one assumes that the lorica illustrated in 
Fig. 6A is not the product of its present occu- 
pant, how does one explain the formation of 
the extensions illustrated in Fig. 6C, since only 
one body attachment point (e) is present? If 
one rejects Andrews’ theory that the present 
occupant (Fig. 6C) is responsible for the lorica 
( d-e ) and its extensions ( c-d and b-c) then 
one must extend Hadzi’s theory to include the 
possibility that that portion of the lorica be- 
tween c-d may have been secreted by a second 
swimmer, and that portion between b-c by a 
third. As improbable as this may at first appear, 
there is some evidence at e (Fig. 6C) to support 
this view. Although it is possible that none of 
the original attachment-point material remains 
(Fig. GA-e ), occasionally (Fig. 6C-e) material 
accumulates whose texture and staining affinity 
appear identical with those of the present body 
attachment material. 
Surely, for those examples in which swim- 
mers have entered old lorica and built exten- 
sions, Hadzi is correct in limiting the count of 
the spiral whorls to those of the original lorica 
and excluding the number of whorls added by 
new swimmers. However, in cases in which the 
