106 
PACIFIC SCIENCE, Vol. XIV, April I960 
Fig. 6. Dolioletta valdiviae, ventral view of gut and 
gonads of gonozooid. M4 to M7, body muscles. 
the belief that D. valdiviae may be a form of 
D. mirabilis. 
From Sewell’s account, it would seem that the 
"position and arrangement of the testis” is the 
only character preventing him from identifying 
his specimens as D. mirabilis. Garstang has dis- 
cussed testis growth and the considerable varia- 
tions in its shape and position resulting from 
differences in the reaction to muscle bands of 
its anterior termination. This applies especially 
in D. denticulatum, but also in D. gegenbauri 
and D. tritonis. The growth of the testis in 
these species, he states, is parietal, which in- 
volves penetration in the body wall between the 
pharyngeal epithelium and successive muscle 
bands: the connections of muscles to the pharyn- 
geal epithelium constitute definite obstacles to 
advance. An obstinate barrier to penetration 
may cause the terminal portion to loop back or 
otherwise alter direction. Of mirabilis, Garstang 
says, the testis begins as a parietal organ while 
in valdiviae it does not. It is possible, therefore, 
that Sewell’s specimens demonstrate an abnor- 
mal condition of the testis, due to some check 
during growth, with a consequent similarity to 
the testis of valdiviae. It would seem, otherwise, 
that the specimens are mirabilis. Nevertheless, 
as Garstang says (p. 219), "An origin of val- 
diviae from mirabilis by a precocious differentia- 
tion of connective tissues impeding the advance 
of the gill septum and the parietal growth of 
the testis is still conceivable.” 
The New Zealand material agrees with Gar- 
stang’s account of D. valdiviae in possessing an 
entire, but narrowed M.6 (Fig. 7) and in the 
coiling of the testis about the intestine. On the 
other hand, Garstang illustrates the dorsal sep- 
tum extending to M.6V2 and the endostyle al- 
most from M.2 to close to M.5, whereas in the 
New Zealand material the septum usually ex- 
tended only to M.6 (once to M.GVa ) and 
the endostyle between M.2 X Z> to M.4V2-M.444 
(Figs. 4, 5). Although these two features bear 
more similarity to D. mirabilis, Garstang has 
pointed out (pp. 204, 217) that variation in 
the extent of both organs does occur and con- 
sequently they may not be important to the 
proposition that the New Zealand material is 
D. valdiviae. 
OLD NURSE STAGE: Garstang (1933) success- 
fully differentiates between the "nurse” stages of 
D. gegenbauri and D. mulleri by comparison of 
the proportional widths of the individual muscle 
bands when the width of each band is expressed 
as a percentage of the total muscle width. He 
finds that in D. gegenbauri, M.3 and M.4 domi- 
nate (with M.3 of slightly higher percentage 
width than M.4), and in D. mulleri, M.4 and 
M.5 are dominant. The distinctions were valid 
TABLE 3 
The Ranges and the Ratios of the Lengths of Body and Tail of 
Oikopleura dioica FROM NEW ZEALAND 
LENGTHS (MM.) 
RATIOS, BODY TO TAIL 
STATION 
NO. 
Body 
Tail 
LENGTH 
Maximum 
Minimum 
Maximum 
Minimum 
Maximum 
Minimum 
Average 
12 
1.1 
0.5 
3.8 
3.4 
3.75 
7.0 
5.4 
32 
0.9 
0.7 
3.9 
3.1 
4.3 
4.4 
4.35 
40 
1.4 
1.0 
5.1 
3.6 
3.7 
3.6 
3.65 
Averages 
3.9 
5.0 
4.5 
