38 T. G. HALLE, MESOZOIC DEPOSITS AND FLORAS OF PATAGONIA AND TIERRA DEL FUEGO. 
divided into two halves which are more or less dissected into 3—6 segments each. 
Segments linear with rapidly contracted, obtuse or rounded apices. Venation con- 
sisting of 5—10 subparallel dichotomizing veins in each segment. 
The specimens referred to this species display a very great variation. Some of 
them might, indeed, be referred not only to a different species but to the genus 
Ginkgo instead of to Baiera. The difference between these two genera is very vague: 
there cannot be pointed out any single character that could serve as a generic distinc- 
tion in all cases, and there exist a number of forms that can be equally well referred 
to both genera. If a narrow angle of the edges and a gradual transition from the 
lamina to the petiole are regarded as characteristics of Baiera, it is at any rate 
certain that the specimens in pl. 4, figs. 23—27, should be referred to that genus. 
The specimen in pl. 5, fig. 1, has a wider basal angle of the edges, but is probably 
a true Baiera. "Typical species belonging to that genus have, indeed, a basal angle 
just as wide and even wider — so, for instance, two leaves figured by SEWARD (1900, 
pl. 9, figs. 5 and 3) as Baiera gracilis and B. gracilis £. Muensteriana resp. — The 
specimen in our pl. 5, fig. 2, appears to have an almost straight base, but this is 
no doubt due to its being split lengthwise and the two halves bent from each other. 
The leaf in pl. 5, fig. 3, has an almost straight base. 
In regard to the dissection of the leaf, all the specimens mentioned above can 
be without hesitation referred to Batiera; but there are, in the genus Ginkgo, forms 
that are very similar in this respect, especially G. sibirica HR. and G. lepida HR. 
The different specimens here referred to Baiera australis display a great variation in 
regard to the number of segments. The leaves shown in pl. 4, figs. 23—25 and pl. 
5, fig. 2, have 3—4 segments in each half; the one in pl. 5, fig. 1, has the greatest 
number, or 6 in the right-hand half, which alone is complete. 'The other specimens 
are intermediate in this respect, one or two of the original three lobes of each half 
being more or less deeply divided. "There is thus a fairly continuous series of forms, 
and although the extreme ones have not much in common, it would seem to be the 
best way to refer them to one species. 
Of all known species, B. australis M”CoY (STIRLING 1892) appears to be the one 
to which the Patagonian form shows the closest resemblance. Unfortunately, I have 
had no opportunity to consult M'Covy's first illustration of this species (STIRLING 1892, 
pl. 1, fig. 2). The second figure of the same species, also given in a report by 
STIRLING (1900, p. 5; pl. 1, fig. 3), shows a leaf of about the same habit as in our 
figs. 23—24, pl. 4, but with somewhat broader and less dissected segments. Two 
other specimens from Victoria figured by SEWARD (Il. c.) under the same name show 
a greater resemblance to the Patagonian form and render it more probable that the 
latter should be referred to B. australis. SEWARD'S two specimens have a small 
number of segments, probably three in each half, agreeing in this respect with the 
less dissected of the present leaves. There is probably a great individual variation 
in this respect; and the specimen figured by STIRLING in 1900 (l. c.) has a consider- 
ably greater number of segments, which are, however, more continuous. On the whole, 
there must be said to exist a very close agreement between the Patagonian form and 
