Africa. It has been possible to examine only one species of this genus. This shows paratracheal parenchyma, often forming conspicuous eyelets' and distinct ripple-marks. In view of these anatomical differences, its affinity to the fossil does not seem to be possible. The next genus, Eperua of Brazil and the Guianas, was not available for comparison. However, Record and Mell (29) report on Eperua falcata as having distinct growth rings while Bancroft (1, 2) reports the presence of ripple-marks in it. By these two features alone, the affinity of the fossil to this genus does not seem to be justified. Prioria copaifera Record and Mell (29) describe as having parenchyma vasicentric and "numerous irregularly spaced tangential and concentric lines". Thus it will be seen that none of the Caesalpinioid woods with normal vertical gum ducts can be profitably compared with the fossil under investigation. Dipterocarpaceae is the only order which now remains to be critically examined. Except Monotes and Marquesia (2) all its genera are known to have vertical gum ducts in their woods. The diagnostic anatomical features of this order are vertical gum ducts, tracheids adjacent to the vessels, vasicentric parenchyma in thin intercepted bands, metatracheal parenchyma in short or or long rows at unequal distance and heterogeneous rays. It will be seen that the fossil under investigation shows all these features in common with the wood of the Dipterocarpaceae. Moreover, the type of pitting found on the different kinds of cells in Dipterocarpaceous wood are also the same as has been recorded for the fossil. In view of these facts it would appear that the fossil G. S. I. Type No. 16502 belongs to the order Dipterocarpaceae. The wood of the Dipterocarpaceae, on account of their commercial importance, have been studied to some extent by various plant anatomists (14, 25, 24, 32, 26, 13, 10). Most of these publications are confined to the woods of different political or geographical regions. Complete anatomical data of various genera and species growing in different countries are not available at present. It is not, therefore, possible to trace the fossil to any particular species. In this connection it may, however, be mentioned here that some seven years back a start was made in this laboratory to collect anatomical data of the various Dipterocarpaceous woods obtained both from India and abroad. This work is not yet complete, but based on the available data, the following tentative classification is suggested. Leaving aside Monotes and Marquesia, the rest of the family can be divided into two main groups. Those which have gum ducts mostly in concentric bands extending over an inch will come under one group. Included in this are the genera, Shorea, Doona, Hopea, Isoptera, Parashorea, Pentacme, Balanocarpus, Dryobalanops and Diopticarpus.* In the other group the gum ducts are always scattered all over the wood--they may be single or in groups of a few. Occasionally one may come across in this group a concentric band of gum ducts but always accompanied by the scattered ones. Seven genera come under this; they are Dipterocarpus, Vateria, Vatica, Anisoptera, Cotylelobium, Stemonoporus,* Pachynocarpus and Monoporandra.* From the description of the fossil recorded in this paper it will be noticed that it belongs to the second group. On making a comparative study of the genera under this group, based on the size and the distribution of vessels and of gum ducts, as well as the shape and the arrangement of rays and their individual cells, it has been found that the fossil shows the greatest affinity to the genus Anisoptera. As to tracing the fossil to a particular species of Anisoptera it could not be done for two reasons. Firstly, our knowledge of the anatomy of the Dipterocarpaceae is not at present complete, and secondly, the unsatisfactory preservation of the fossil does not allow the required measurements that are necessary for separating the species. (3) Comparison with those previously recorded. (a) INDIA. In 1916 a fossil wood from the Irrawaddy (Tertiary) series of Burma was identified by Holden (18, 36) as Dipterocarpoxylon burmense. Later Gupta (16) re-examined and re-named it as Irrawadioxylon burmense. In a previous paper the author (6) has expressed his doubt regarding Gupta's identification and has also pointed out the affinity of Holden's original specimen to Glutoxylon assamicum. In any case, it is now certain that this fossil from Burma has been wrongly named as Dipterocarpoxylon. Sen (35) reported another fossil wood from Bengal and called it Dipterocarpoxylon after Holden. Since Holden's specimen was not correctly identi * Due to divergence of opinion amongst systematic botanists, the generic position of these four is still uncertain. fied, it appears to be justified to conclude that Sen's specimen is not a true Dipterocarpoxylon. In 1935 Gupta (17) reported another fossil wood from the Tertiary formations of Burma as Dipterocarpoxylon holdeni. The photomicrographs included in his paper did not show much affinity to the woods of the living Dipterocarpaceae. At my request Professor B. Sahni of Lucknow University has been kind enough to send me the slides of this fossil. I have examined them and am of opinion that Dipterocarpoxylon holdeni has not been correctly identified. To start with, it does not possess any vertical gum ducts. Nor does it show similarity to the woods of Dipterocarpaceae in size and distribution of vessels, of parenchyma and of rays. In the face of all these differences there does not appear to be any justification for taking it as one of the Dipterocarpoxylon.* Recently Rode (33) described a fossil dicotyledonous wood from the Deccan Intertrappean beds and named it as Dryoxylon mohgaænse. He did not attempt to show affinity of the fossil to living trees but simply recorded some of his observations. From what he has given in his paper it appears that the fossil from the Intertrappean beds does not belong to the Dipterocarpaceae nor to the fossil wood from the Garo Hills. From the above it would be seen that the fossil under investigation is the first specimen in India to be recorded as belonging to the family Dipterocarpaceae. (b) OUTSIDE INDIA. Different workers have used the name Dipterocarpoxylon in different senses and it seems here desirable to clear up this point before an enumeration is made of the various Dipterocarpoxyla that have so far been recorded from outside India. Kräusel (20, 21, 22, 23), Edwards (11) and Bancroft (1, 2) are of opinion that the name Dipterocarpoxylon should be used to include all fossil woods of the Dipterocarpaceae, while Den Berger (8, 9), Pfeiffer and Van Heuren (27) have gone further and divided the family into different genera or sub-groups and named them accordingly. In the opinion of the latter three Dipterocarpoxylon includes only two genera, Dipterocarpus and Anisoptera. *I understand Dr. Gupta is soon going to re-examine this fossil critically, and if possible, determine its correct systematic position. |