PAEKA DECIPIBNS. 15 



cell-wall. These bodies are represented in fig. 2 b, the figures given having been taken 

 as fair average examples ot all those found. As the result of an examination of the 

 material first submitted to me, I came to the conclusion that these bodies have an average 

 diameter of 34"6 /'., but that sulHcient allowance for the thickness of a cell-wall, as repre- 

 sented by the carbonaceous investment, would make their original dimensions approx- 

 imate to 40 /'. Later and more extended, measurements confirm this result. Comparing 

 with modern types, I find these bodies to be a little larger than the spores of Lycopodium 

 (34 /(.), much larger than the microspores of SelagineUa (28 /'.), and small as compared 

 with the macrospores of extinct Prolosalvania or modern rhizocarps. 



There were also found two small oval cells (fig. 2 a) possessing the remains of an 

 outer cell-wall as a carbonized crust. They measured 15 //. in diameter. No other simi- 

 lar bodies were found. This might be due to their more perishable nature. Both of 

 these bodies are evidently of the nature of spores, and it would seem justifiable to regard 

 them as macrospores and microspores, a view justified by their difference in size, state of 

 preservation and relative number. This would therefore show that both kinds of spores 

 were produced in the same sporocarp as in Marsilia and Pilularia. 



An additional fact of very considerable importance seems to be well established by the 

 material examined. In the slides prepared by Sir William Dawson, who specially drew 

 my attention to them, there are found a large number of rather curious, apparently stellate 

 bodies of complex structure. Sir William Dawson seemed inclined to regard them as 

 remnants of spore clusters. My own impression, as I first examined them, was that they 

 represented the walls of empty sx^oraugia. Upon more critical examination, and after an 

 inspection of additional material, it became evident that neither view could be maintained. 



It was found that each body (fig. 1) contained a central carbonized mass, probably 

 the remains of a more highly cutinized structure, about which were disposed cells of very 

 variable form and size, often showing a more or less distinctly radial disposition. Careful 

 focussing also showed that the smaller ones (/) were spherical, while the larger (a) 

 were nearly or quite flat. An examination of all I could find showed that there was a 

 total absence of constancy in size and structural detail ; and if sporangia, this variation 

 could only be accounted for on the ground of different degrees of development, a condi- 

 tion most unlikely to be found among sporangia from the same sporocarp. Their dimin- 

 utive size as compared with the Parka discs (sporangia) would also be opposed to this. 



In fig. 1, comparing the series of figures, it will be seen that from /to a there is evi- 

 dent a more or less gradual development. These figures were taken without special 

 selection as examples of the many structures of this kind found, while scA^eral similar 

 bodies much larger than a were also observed. The possible connection between fig. 1 / 

 and fig. 2 6 is somewhat obvious. If we regard/ as a developing condition of b, and that 

 from /to a we have different stages of growth in similar bodies, I think the relation is 

 made clear. This would, therefore, make the bodies a, b, c, d, e, f prothalli in different 

 stages of growth, a view which is well supported by their form and structure. 



The data thus gathered seems to strongly indicate that Parka is an aquatic rhizocarp 

 allied to Pilularia. In the earlier part of the correspondence on this subject Mr. Reid 

 inclined to this view, but later thought there were possible affinities with Marsilia. I 

 think we may consider, however, that the narrow leaves, the sessile fruit and the com- 

 position of the sporocarps make the relation much nearer the former than the latter. 



