504 
Fig. 3. Diagram of a cut gall (not to scale), m, 
mite; e, epidermal hair; x, xylem strand; pm, pouch 
meristem; o, orifice of mite cavity; p, periderm. 
ating from the embedded pedicel, and giving 
the appearance of the spiky "coat” of a chest- 
nut fruit. 
The mite-containing cavities are undoubtedly 
the most remarkable feature of the galls. The 
mouths of the cavities are irregularly creased 
(Fig. 3) and surrounded by stout epidermal 
hairs which are also found in clumps on other 
regions of the gall surface. The lining of each 
mite cavity is in fact a sac-like growth centre — 
a kind of pouch meristem analogous to the 
shoot growing point of higher plants (Figs. 
3 and 4) . 
Mitotic divisions are found most often within 
a zone about three cell layers from the cavity. 
The histological methods used in this investi- 
gation were selected to provide a general ana- 
tomical picture of the gall and a cytological 
analysis of mitotic rates was not intended. 
What clearly emerges is the fact that peren- 
nial growth of the mite-induced galls of Ho- 
heria sexstylosa is dependent on the regular 
activity of the "pouch meristems.” 
Only at the orifice of the cavities are the 
epidermal hairs differentiated (Fig. 4), no 
doubt giving shelter to the colonies of mites 
below. To the interior of the gall and close to 
the middle region of the "pouch meristem,” 
xylem elements are seen to differentiate and 
PACIFIC SCIENCE, Vol. XIX, October 1965 
lead backwards in an anastomosing system with 
vascular traces from other mite cavities to unite 
in the central stalk (Fig. 3). A coherent and 
efficient conducting system is thus constructed 
and the gall is able to flourish to the detriment 
of the rest of the twig ( Fig. 5 ) . 
To the superficial glance the appearance of 
Hoheria sexstylosa galls, whether borne on 
leaves or on flower stalks, is of an irregular 
spherical or conical shape, furrowed by indefi- 
nite crevices. But the underlying histogenetic 
processes of periderm formation, cavity growth, 
and vascularization are remarkably regular. No 
essentially different cellular components, except 
the mites, distinguish the galls from the normal 
shoot tissues, but the growth pattern is unique. 
DISCUSSION 
So little is known of the biology of Eriophyes 
hoheriae that it is premature to consider in what 
manner the gall mites may influence growth of 
galls in Hoheria sexstylosa. Whether the mites 
provide a stimulus in the mechanical act of 
feeding, or by the liberation of growth sub- 
stances or enzymes, remains to be investigated. 
The preliminary attempts to induce gall 
formation (described in the section on Meth- 
ods and Materials) proved ineffective, possibly 
because the winter season was an unfavourable 
time. However, other workers have reported 
promising results from experiments in which 
galls were artificially induced by extracts from 
various insects (Bloch, 1954; Boysen- Jensen, 
1948; Braun, 1959). 
Although the onset of galls in young leaves 
has been noted at the same time as the galling 
of flower buds in summer, observation of sev- 
eral trees over a number of years has failed to 
disclose galling of leaves in spring when there 
is a mass unfolding of leaf buds. The lack of 
success in the attempted transmission of mite 
infection to young leaves in the glasshouse in 
July, though not very cogent evidence, tends 
to support the view that new galls arise pre- 
dominantly in late summer. 
The production of galls on leaves would ap- 
pear to be less favourable to the maintenance of 
the mite population than the induction of galls 
on flower buds, which give rise to longer lived 
galls. Badly galled leaves frequently absciss in 
