334 
THE GEOLOGIST. 
walls of tlic pipes in the magncsian limestone seems to afford ano- 
ther argument in favour of this theory, for this effect is undoubtedly 
due to chemieal action of some kind ; though some may hold that 
the decomposition of the surface may have resulted aftei' the forma- 
tion of the pipe by a very different agent ; and this is certainly 
possible, though it must be evidjnt at the same time that whatever 
decomposed the limestone forming the surface of the pipe, had suffi- 
cient joower to be the primary cause of their formation ; for if the 
application of the decomposing agent effected the results we see in a 
certain amount of time, it is not to be doubted that by increasing the 
period of apjjhcation, so would we increase its results ; so that 
having here a power competent to originate the phenomena of sand- 
pipes, it seems more philosophical to credit it with their consumma- 
tion than to call in the aid of another power whose capabilities even 
to originate them is almost more than questionable. 
But though the chemical theory of the origin of sand- and gravel- 
pipes is more satisfactory than the mechanical theory of Mr. 
Trimmer, or than any conceivable theory of a mechanical nature, and 
though I have little doubt myself but that they have really originated 
by chemical action of some kind, yet there are one or two points 
connected with this theory which seem difficult as yet to explain. 
For instance, it is not easy to understand how the water contained in 
the stratum overlying the limestone could be so extra-charged vrith 
carbonic acid as to possess erosive power enough to excavate the 
pipes. It seems plain that more than the usual quantity of carbonic 
acid would be requu-ed, or sand-pipes would be of more common 
occurrence where limestone surfaces are exposed to the reach of 
rain-water ; and if the roots of vegetables supplied the extra quantity, 
as Sir Charles Lyell suggests, I see no reason why they should not be 
found Avhere limestones lie immediately beneath the turf, which I 
believe is never the case in any district where sand-pipes occur. It 
is possible, however, that rain-water may derive from vegetable 
matter, as in the case of an overlying morass, an addition to its usual 
per centage of carbonic acid ; and the fact of the remains of so many 
small roots being found at a depth of three or four feet in the core 
of clay of some of the pipes I have described, may, perhaps, be con- 
sidered as rather indicative of something of the kind. It is also 
possible that it may derive additions from animal remains imbedded 
in the overlying alluvium, as Mr. Prestwich suggests ; and the 
absence of all remains of this kind in the alluvium in question is no 
proof to the contrary, for this supposition necessarily includes their 
destruction in the derivation of their carbonic acid. But still, not- 
withstanding the possibility of such supplies, I must confess that to 
mc it seems probable that we are still ignorant of the true source of 
the erosive agent ; neither do I see any reason for supposing that 
carbonic was the only acid employed, and that by it alone were the 
pipes eroded. 
i^Lnother difficulty is the special application of the acid or chemical 
agent to the particular spots occupied by the pipes. If we suppose 
