54 
REPORT— 1847. 
tbe solid envelope, would form wbnt may be described as inverted hollowj is 
its lower surface. Let us first take the hypothesia of a previously existing 
solid central nucleus. The thicker undisturbed portions of the solid shd 
would become united with it by continued solidification, before the more 
disturbed parts would become so, and thus the hollows above mentioDed 
Would become cavities more or less detcriiiinatc, bounded in all directions by 
solid rocks, and filled with fluid matter. Their horizontal dimensions tnigiil 
in the first instance be very large, but continued refrigeration M'ould not only 
tend to diminish thens, hut also tfi divide each cavity into a uumborofsinnllff 
ones, by a process exactly similar to that by which fijcse laiger cnTiiiwvtre 
themselves formed from the general fluid mass, the position of each minof 
cavity being determined by the points in tlie roof of the larger one white 
previous fractures had left tin? most perfect volcanic vents. By a coiitioui* 
tjon of this process it is obvious tluit the superficial crust of our globe niwt 
at length arrive at wliat I conceive, for reasons above o^eigned (avtS), lobs 
that of a solid mass containing numerous cavities filW 
with fluid incnndesoent matter, mid either entirely insulated or perhaps com- 
mumcatmg in some cases by olistructi'U t-lmnncls. 
IP above explanation would manifestly be equally applicable whatever may 
nave been the process by which the globe may be conceived to have arrived 
at the state above supposed previous to tlm commencement of the superfickl 
incrustation,—that in which it consisted ofaoli,! central nucleus witha fluid 
envelope. 1 oisson seems to have thought it probable that the earth had 
long lost all traces ol its original heat, and that the present U'rrcstrial tern- 
of the solar system having pmei 
in thflf wi *1*^®*^ «'Bieh the temperature was much higher than 
the ifvnn T"' °<^cupi(.s, It would reifuire an extension o 
admittSe its nrni”? P'^riiaps considerably iiicrea-se the difficultyof 
have beem ^“P^^rficitU portions of the earth to 
ceTvobL I*™-'--' by •!". or oL - 
If we tak#. »h.. u ^ *’*’aso»iinp will remuiii the same- 
tho surface th«> *'^'^** eolidifioation having commcDced at 
account for’the >*^a^onmg fart. l(i) will still be applicable to 
mentioned, but we have herp^nn^ those above 
their floors Tlif. .m • existing solid nucleus to for® 
as great in this case as in th/i- the crust will beatM 
floor beneath the part of the 1101 * 1 *^'^’ tor the fbrmalioD of a sobd 
must assume that a rertain fluidity is thus maintained, we 
vent the generation m.d ^“1" t ^ « be sufficient to pr^ 
effect of prmurc iu -nreventim^ ^ .^.V®'■‘*P««f«» though, by hypoiJie>«i tlw 
we are considering th^an in tbJfc “‘rrst be much icis« in the 
--.imptiou of a limiiZneath wbl I" Admitting thn 
tiou of elastic gases it U ninriT ^ .i ^ pr’r’Muro would prevent the form»- 
Holidify when its U-inneratu^^ h ‘ 
lidifiration of the imrtioiis firrmST fediioed. viz. when the so- 
tbat limit. At greater denths l ^ of the cavity had reached 
tho causes hitherto specifiSj nud wruhtfh^T interrupted bj 
causes should operate) urollurp (unless otheraud unknown 
hmit Just mentioned the soHdiv mass, ivhile aboi'e the 
in a state of HuidUy ^ -nterrupted by portions of the mais 
Accordi„g .a. „.oJ„ hero (thatiu 
