Penrose. J PHOSPHATES OF ENGLAND. 89 
matter. Granular patches of a deep-red color are sometimes scattered throughout 
the lighter-colored portions. Spicules occur in many sections, presenting some of 
the most characteristic forms of spongo-spicules; as, for example, hexaradiate, triradi- 
ate, hamate, sinuate, and connecting forms. These spicules are frequently grouped 
together in a manner which seems to indicate that they cannot have been washed in 
from the seabed during fossilization. Globular bodies 4( ' )ir inch in diameter are nu- 
merous ; they seem to be gemmules. Pohjcistina and Xanthidia occur in some sections. 
With rjolarized light the sections appear distinctly cryptocrystalline, presenting an 
appearance very nearly resembling that of chalk Hints when examined in the same 
way. A very curious phenomenon may be alluded to here. A number of small circles 
may be seen in some sections, each of which is marked by a black cross, the arms of 
which radiate from the center to the circumference. On turning the analyzer the cross 
revolves and, when the analyzer has been turned round 90°, is replaced by a comple- 
mentarily illuminated cross. The explanation of these appearances seems to be as 
follows: Small Globigerina shells and other similar spaces occur in the nodules, into 
which the crystalline apatite, which was diffused throughout the fossil, has pene- 
trated and crystallized inwards from their walls to their centers, thus forming a 
radiating mass of crystals. It is well known that crystals arranged in this way will 
produce the phenomena described. 
Mr. Sollas 1 thinks that many of the nodules of the Upper Greensand 
are phosphatized sponges. Others he considers to be "phosphatized 
animal matter decomposed so far as to have lost all traces of its orig- 
inal structure before mineralization." He found fish scales and bones 
in many of these nodules, and therefore concludes that the animal mat- 
ter was sometimes derived from small fish. He does not seem to take 
iuto consideration that the scales and bones may have been buried in 
a matrix of calcareous matter, and that this substance, whether it was 
marl or limestone, may have been phosphatized, thus forming phos- 
phatic masses, which, of course, would contain the same fossils as the 
original calcareous substance from which they were formed. 
The phosphate bed of the Upper Greensand varies considerably, not 
only in the quantity of phosphatic nodules, but also in the chemical 
composition of the individual nodules. It is often found that two places 
may be equally rich in the quantity of nodules, while the content of 
phosphoric acid in them may be widely different. 
All through the counties of Dorset, Somerset, Wilts, and Devon the 
nodules are very much more siliceous and less abundant than the nod- 
ules of Cambridgeshire and Bedfordshire. 2 Besides the variability io 
the phosphatic richness of the bed, it is also sometimes very variable in 
its mode of occurrence. At times it will cover many square miles con- 
tinuously, while at others it occurs iu pockets in the surface of tbe 
Gault. At other times, according to Mr. Fisher, the bed shows signs 
of contortion, as indicated in Fig. 34. It will be seen, by examining the 
analyses given beyond, that the Greensand matrix of the phosphaiic 
nodules varies also very much in its content of phosphoric acid. The 
1 Quart. Jour. Geol. Soc, vol. 28, 1872, p. 398. 
2 L. Jenyns: Geol. Mag., London, 1866. 
(563) 
