fVood partly petrified by Carbonate of Lime. 209 



ference, but more frequently only insulated portions of about one tenth of an 

 inch in diameter, in which also the structure is well preserved. The direction of 

 the rays, which pass through these portions, shows that they are, with scarcely 

 any exception, in the same position as when the stem was entire, although they 

 are now surrounded by mineral matter (carbonate of lime), in which scarcely 

 a trace of the structure of the plant remains. 



From this condition of the fossil, it appears, that as in the former instances, the progress of petri- 

 faction commenced at a number of separate points and was afterwards interrupted, but in the present 

 case, the whole of the remaining portion of the wood appears to have been entirely destroyed by 

 decay, which occurred after the interruption, and no renewal of the process took place, while any 

 of the other parts of the wood existed. 



The fossil wood from Linnel Braes, described and figured by the same 

 author (PI, 3), is also a case where parts only of the wood have been petrified 

 and the remainder entirely decomposed by subsequent decay, but the petrified 

 parts are in their places, and the intervening space is filled up by crystallized 

 carbonate of lime. 



Such are the appearances shown by sections made in a transverse direction, 

 or at right angles to the axis of the growth of the wood. In the examination 

 by means of longitudinal sections, we shall find the following differences be- 

 tween them. 



In the recent wood from the Roman aqueduct, the carbonate of lime runs, 

 as I have stated, through the whole thickness of the specimen, the petrified 

 portions appearing to be in long columns. In the Antigua specimen, the well 

 preserved portions are not, in any of the instances I have seen, more than three 

 eighths of an inch in the longitudinal section, and their form is irregular. 



In the Allen Bank fossil, the insulated petrified portions are of the same 

 size in the longitudinal as in the transverse sections, showing that they are of 

 spherical form. 



In the last named instances it appears to me, that we must conclude, that the whole of the wood 

 was penetrated by a saturated solution of the mineral matter, and that the solution was in the state 

 at which crystallization was about to commence. In solutions in such a state, we know that cry- 

 stallization frequently begins simultaneously at a number of separate points, round which the cry- 

 stalline matter continues to aggregate. In the Allen Bank fossil, at the external part of the section, 

 this has evidently taken place, as an examination with a magnifying glass shows, that although the 

 structure of the wood is preserved undisturbed, yet that the petrifaction is there made up by the 

 contact of a number of small spheres. The progress of petrifaction, however, appears to have 

 stopped before it had mineralized the whole of these specimens, and this may have been caused by 

 an accession of water diluting the solution, below the point at which crystallization would go on. 

 The variation in the quantity of rain falling at different periods, or in different seasons^ might 

 occasion such a result. After this, as I have before remarked, the unchanged portion of the wood 

 was subject to decay. In the case of the Antigua specimen, before the process of petrifaction 

 VOL. V. SECOND SERIES. 2 E 



