Dr. F. A. Bather—Collodion Imprints of Fossils. 439 
Fossil wood is readily studied by this method, especially when 
silicified. In this case the broken surfaces generally are transverse, 
radial, and tangential, or at any rate such surfaces can be readily 
exposed by a few blows of the hammer. Collodion imprints of these 
are almost equal to thin sections taken in the same directions, and, in 
coniferous wood for example, show the medullary rays, the tracheids, 
and the bordered pits with remarkable distinctness. Imprints from 
a silicified fragment of COupressinoxylon, originally described by 
Conwentz in 1891, show every cell in the clearest manner even. under 
high powers of the microscope. Wood from Brown Coal, if well 
preserved, also furnishes good imprints, especially of surfaces in 
a longitudinal direction. 
Plant-remains preserved in the finer sediments of chemical origin, 
such as calcareous tufa, the Mesozoic silicious rocks of Franz Josef’s 
Land, and some clay ironstones, have also yielded good results. Thus 
a piece of a Conifer from the tufa of Benestad in Scania gave imprints 
showing all the points above mentioned. Leaves in tufa have not 
always been completely embedded before decay set.in, and may not 
always furnish serviceable preparations. A very successful imprint, 
was, however, made from a leaf in the Scanian tufa, referred by 
C. Kurck in 1901 to Zussilago farfara. This preparation showed the 
outlines of the epidermis-cells, the hair-scars, the stomata, and the 
peculiar streakiness of the outer cell-wall, even under a magnification 
of 500 diameters and upwards (Fig. 1). Thus Kurck’s determination 
was fully confirmed. he imprints taken from these tufaceous fossils 
must sometimes be cleaned by acid from the adherent calcareous 
particles. 
As regards leaves in the silicions rock from Franz Josef’s Land, it 
is enough to mention that Professor Nathorst now publishes photo- 
eraphs of films showing the structure of the epidermis-cells and the 
stomata in Desmiophyllum, features that Solms Laubach, who studied 
the same specimens in 1904 with the aid of thin sections, had not been 
able to discern. 
The advantages of the method as applied to fossil plants may now 
be summed up. It is not claimed that collodion imprints can 
altogether supersede thin sections, although, as the last instance 
shows, there may be cases in which they reveal even more than the 
sections. But there are many cases in which either time or the state 
of preservation of the specimen, or perhaps its owner, do not permit of 
thin sections being made; films then form an admirable substitute. 
It may often be a saving of time to work over a collection rapidly by 
this means, sorting fragments into their genera, and selecting for 
further treatment those that seem to promise best results. Why, it 
may be asked, not put the specimens directly under the microscope ? 
Because in some cases they are much too large to be so handled, and 
in any case the structures cannot so easily be seen by reflected light. 
The last reason reminds one how readily these films can be photo- 
graphed, although even the excellent photographs reproduced by 
Professor Nathorst cannot, as he points out, reveal the structure so 
well as the films themselves when studied with a varying focus. But 
this suggests another advantage: the films are easily made, and can 
