ANATOMY OF THE JULIANIACEX. 13] 
the relationships of Julianiaceze and J uglandaceze, states that the latter are resini- 
ferous; the resin is here, however, contained in the cells and not in special secretory 
receptacles, as in Julianiacesw and Anacardiacem (cf. Engler, in Natürl. Pflanzenfam. 
iii. 1, p. 21). 
The anatomy gives evidence of a far closer relationship to Anacardiacee (Hemsley, 
p.191). In the first place, the resin-canals mentioned above are an indication of close 
affinity ; they occur in just the same position as in the Anacardiacee, being invariably 
found in the phloem throughout the plant. In the Julianiacez resin-canals are also 
present in the primary cortex in all cases; this is not the rule in the Anacardiace:, but 
canals are found in this tissue in some Anacardiaceous genera. "The presence of canals in 
the pith is not an absolutely constant featuve in either Order. The mode of development 
of these secretory organs is the same in the two cases. A further point of resemblance 
between Anacardiacez and Julianiaceze is to be found in the peculiar external glands of 
the latter, which are very much like those of a Dobinea for example *. Reference to 
Solereder and to the section of this paper dealing with the structure of the stem will 
show that the wood is very similar in the two Orders. ‘The place of development of the 
cork is also the same in both cases. In fact, there is so much resemblance in the 
anatomy between the two Orders that it is difficult or almost impossible to separate them 
on anatomical grounds,—the more so as the anatomy of the leaf in the Anacardiacez is 
but very imperfectly known. It is possible that when we know more about the latter 
point some distinguishing character may become evident; perhaps the almost complete 
absence of sclerenchyma in the vegetative organs of the Julianiaceze and the prevalence 
of clustered crystals may be found to be of value in this respect. 
It still remains to consider the Cupulifere. Anatomical structure does not encourage 
the view of a close affinity of the Julianiaceze with this Order. The Cupuliferze have no 
internal secretory elements; the structure of the wood is very different in the two cases ; 
the pericycle contains a sclerenchymatous ring in the Cupulifere, &e. On the other 
hand, there are a few points of resemblance (absence of subsidiary cells to the stomata, 
occurrence of glandular hairs which are in some eases a little like those of the 
Julianiaceze, subepidermal cork-development), but these are of little importance in view 
of the absence of the main characters. 
The previous paragraphs may be briefly summarised by stating that the Julianiaceæ 
in their anatomical structure show a most marked affinity to the Anacardiacese,—so 
marked, indeed, that it is difficult to hold the two Orders distinct from this point of 
View. Yet in the external morphology Mr. Hemsley finds so many striking points of 
difference that he comes to the conclusion that the balance is in favour of a Cupuliferous 
affinity of the Julianiacese +. It seems, however, that the Julianiacez in their external 
morphology show a decided correspondence neither to the Cupuliferze ‘nor to the 
Anacardiaceæ. In coming to his conclusion Mr. Hemsley has perhaps laid too much 
weight on outward resemblance of the inflorescence, &c., to corresponding structures in 
™ Cf. Radlkofer, in Sitz.-Ber. Akad. der Wissensch. München, Bd. xviii. 1888, p. 383; also Solereder, loc. cit 
pp. 278, 279, 
t Hemsley, p. 192, 2 
x 
