Yol. 65.] plant -containing nodules prom japan. 199 



show an opaque matrix, formed of the very fine-grained silt that 

 was accumulating to form the roof and was included in the con- 

 cretions growing round the plant-fragments as centres. This dis- 

 tinction is borne out by the chemical analyses of these two types 

 of plant-containing nodules (see Stopes & Watson, op. cit. pp. 193 & 

 196, for full details, and this paper, p. 200, for single analyses), as 

 well as the physical conditions under which they appear to have 

 been formed. Knowing, then, that the plant-containing nodules of 

 Hokkaido were formed much farther from the protected quiet of a 

 coal-deposit, and that in fact they were almost at the shore in 

 some cases at any rate, it is not surprising to find that they are 

 very much more granular than the Palaeozoic ' roof-nodules.' 



A photograph of a portion of the matrix of a Japanese nodule is 

 shown in PI. IX, fig. 1, where the numerous and irregular grains 

 in the matrix can be seen all through it. Many of these grains 

 are quartz-crystals — things practically unknown in the ' roof- 

 nodules.' Among the granules in some specimens are compound 

 silicates, some of which are brilliantly coloured red or green under 

 normal light. There are also calcite-crystals, some of large size, 

 and a number of opaque granules. The number and size of these 

 granules and crystals vary considerably in the different nodules ;, 

 some being coarse throughout, some exhibiting coarse - grained 

 bands or layers, and some being almost as fine-grained as the 

 Palaeozoic roof-nodules. 



In most of the nodules also are shells, some of minute size, others 

 merely portions of various large moll-usca and ammonites. The 

 number of the shells in a nodule is also very various, but some 

 are to be found in nearly every case. These are often mixed with 

 fragments of plant-tissue, as is seen in PL IX, fig. 2, s. 



Plant-fragments are found with well-preserved tissues in many 

 of the nodules, both in those of very large and in those of average 

 size. In a number of cases the fragments are isolated, a single 

 stem or twig in a single nodule, as is the case in the typical 

 Carboniferous ' roof-nodule.' But in most cases two or more 

 fragments lie together, and in a few of the nodules the plant-tissues 

 lie so thickly massed, and contain portions of such a variety of 

 organs that the nodule may be said to approximate very closely to 

 a true ' coal-ball.' Such a case is illustrated in PI. IX, fig. 3, 

 which presents the same appearance as a 'coal-ball' with its crowded 

 mass of debris. 



In most cases, however, even where there are numerous frag- 

 ments from several plants in the same nodule, there is a quantity of 

 matrix round and between them, a point in which these nodules 

 differ from ' coal-balls.' 



In the co-existence of plant and animal remains in the same 

 nodule, the Japanese nodules agree with the ' roof-nodules ' of the 

 European Carboniferous Period ; while in.the presence of numerous 

 small fragments of plant-debris of many kinds in the same nodule 

 they agree with the ' coal-balls.' 



