CARYOPHYLLACEAE. 



PI. VI, fig. 19. 



A fragment of the testa of a large seed is the only evidence we have for this family 

 in the Reuverian stage. The seed appears to have germinated, and in germinating has 

 turned inside out, so that the concave surface shown in fig. 19 is really the outside of 

 the seed. 



Tubercles in concentric rows, prominent towards the circumference, becoming 

 small and crowded towards the centre, bases interlocking and foliate. 



Length (of fragment) 3.25 mm. Reuver. 



Though the testa is much decayed it shows here and there the foliate sutures 

 between the tuberculate cells, so characteristic of the Caryophyllaceae. 



NYMPHAEACEAE. 



NELUMBIUM MINIMUM Sp. nov. 

 PI. VI, figs. 20 a, b. 



Torus (mancus) minimus, carpellis saltem 22; carpella minuta, arctius con* 

 ferta, ovata, apice conspicuo mammiliformia, striata, basi haud intrusa. 



Part of the middle of a torus less than 5 mm. across, in which are imbedded 

 22 carpels ; carpels minute closely packed and often facetted or dented by mutual pres* 

 sure, ovate, apex prominent mammilate and striate, base not indented. 



Breadth of torus 4.5 mm. Carpel, length 1.5 mm., breadth 1.0 mm. Swalmen. 



This unique specimen is a fragment of a torus, with the margin only preserved at 

 one point; its lower surface is torn away so as to expose the base of the carpels. It shows 

 22 carpels more or less perfect; the curvature of its upper surface and the radiation of the 

 nuts suggest that the total width may have been about 10 mm. and the number of 

 carpels perhaps 40. The exceeding minuteness of this specimen makes it difficult at first to 

 believe that it can belong to the same genus as the living Nelumbium. It shows however 

 the characteristic carpels sunk in a flattened torus; its nucules greatly resemble those of 

 N. speciosum, but are much smaller, more in number, without the basal pit, and have a 

 more conspicuously mammilate apex. The channel of the large funicle passes out of the 

 carpel on a level with the upper surface of the torus. There is nothing to suggest unripe* 

 ness in this specimen; the carpels seem fully developed and hard, and are being set free 

 by the breaking up of the torus. They appear to be crustaceous and fragile; and this may 

 be one reason why no detached carpels have been found in the same deposit. 



BRASENIA PELTATA Pursh. 



PL VI, figs. 21-24. 



Seed not well preserved, more oblong and somewhat broader, but apparently 

 inseparable from the living species. Its testa shows the characteristic triradiate cells and 



85 



