Scott and Maslen . — The Strtccture of T rigono car pus. 95 
be explained as a result of obliquity of the plane of section, which will pro- 
duce a deceptive appearance of an annular ridge surrounding the upper part 
of the seed, as will be more fully explained in a later part of this paper. 
Some of the Hooker and Binney slides show the tracheal elements 
of the nucellus in a well-preserved condition, and these are described and 
figured in their paper. All the Hooker and Binney slides are longitudinal 
sections. 
The next description of the structure of Trigonocarpus was by 
Williamson in 1877 under the name of Trigonocarpon olivaeforme. He 
described and figured a considerable number of vertical and transverse 
sections of good specimens from the Oldham deposits, at about the same 
geological horizon as that from which Mr. Binney obtained his examples. 
The specimens were cut into series of many longitudinal and transverse 
slices which are figured in natural size. The most instructive of these is 
a series of transverse sections (see Williamson’s paper, Figs. 100-8), which 
start from the lower portion of the micropylar tube, and traverse successively 
lower zones of the seed, and which were very useful for comparison with 
the longitudinal sections which had previously been figured by Hooker and 
Binney. With the aid of these transverse sections, Williamson gave the 
first clear explanation of the origin of the ordinary triangular internal casts. 
He also described in detail the microscopic structure of the inner hard 
layer of the seed-coat which he calls the endotesta ( = our sclerotesta) and 
gives a good figure. 
In 1900 Mr. George Wild published a paper : ‘ On New and Interesting 
Features in Trigonocarpon olivaeforme 2 ’ from observations based on 
a number of new slides which have since come into Dr. Scott’s possession, 
and some of which have been made use of in the preparation of our paper. 
From a study of thirteen consecutive transverse sections, Wild was able to 
show (1) that the micropylar beak was very much longer than had 
previously been suspected, being nearly or quite equal in length to the 
body of the ‘ fruit ’ itself ; (2) that the micropylar beak retains its character- 
istic triangular cavity nearly up to the apex, and (3) that the long micropyle 
was apparently winged, the soft (outer) layer of the testa being developed 
in a flattened form on both sides of the central harder tissue. This 
interesting feature will be discussed later on in this paper. 
Wild also emphasized the fact of the frequent association of Trigono- 
carpus with Myeloxylon , which is now known to be the petiole of Medullosa . 
The block from which Wild’s slides were cut abounds in petioles, roots and 
other parts belonging to Medullosa , and most of the other preparations 
which we have seen agree in this respect. In Wild’s slides the limiting 
layers of the soft testa of the seed are often well preserved, and Wild 
pointed out the close similarity in structure between these and the outer 
1 Williamson (’77). a Wild (’00). 
