FOURTH UCTTWl ON ULAOIKHN, [1842. 



of tho glador, and that it in not unfroquontly ontiroly 

 doMtroyod in one, part of tho glaoior, to bo renewed in n totally 

 d.iieront direction iu another, A molecule of ice is a passive 

 and NtruoturoloMN a unit an a molecule of water, to far ad it IUIH 

 not that Ntructuro improved by Momothing external at the time, 

 hike tho walor in die river, wyriatlN Ninverd ,.,,,- another, and 

 MM hi 1 10 iniMluhon for (ho MIUH 



Kow WOIH!M will MM (Hoe to ihew how intimately what 1 havn 

 dUted ii connected with tho lirni nidiiuonU of a thoory of glacier 

 motion, wluoli 1 ondoavottrod to Nkctolt in my IUH(. lottor, and 

 ilio truth of whidi all that 1 havo ninro MOOH has Imdrd .r t ;iil\ 

 to ..'ni.in. Th<< oontro of tho glacior ntroant in urgod onwafdl 

 by proHMuro from abovo (how oausod wo nhall immodiatoly ron- 

 id' 1 ! 1 ), which in tlioro nHiMt.od IOMH llian .n ilt> id< . and liolloiu, 

 nwiny; h |.ho rumpamtivo ubmMu'o of iVirlion. Thr lai.-nil purl.'; 

 ilt dl'IVggOll oii\vaid'. liy lli- UK. lion t>!' (hr tMMilrr. and inovr 



H!MO, but it in quito oompatiblo with tin !! i ->i ..nnilmd mntion, 

 tliut tho bottom of the glaoior nhould romain Iroxon to iu* bod, 

 a Nomo writorn havo Ntipponod to bo ihr rase, i hough I am fur 

 from aborting this to bo tho faot, or ovon NuppoHtng it probablo. 

 Why, thon, aro tho tinNuroN gtnorally wtfmt, and also \\IUMV . 

 glaoior i mont rogular, nimply tmnsn-rsr. and not convex 

 towardn tho lower oxtromity ? Tho tlrnt of thom^ quoNt.ionn 

 had alwn\ nil latol^ appoaral to mo a MoriouM dillioulty, 'I'lu- 

 /.'/.-/ i.i.ji.-d iu ill.- ...-i-iMid, 1-ond'Mn-d \\itli iln> poNJiivc owtainly 



tluit- tho OOntl'O of a j-danrr nin\r. I'a'.i.M ill. in H-. ides, in (lie 

 ralio IVripu-nllv ol' .> l. .", .lu-\\ . dial an answt-r inusf ! lonnd, 



and, thoroforo, that it olVorn no innurmountablo objootlon. Tho 

 explanation tN to bo nought in tho oontinuftlly varying condition 

 >l tho glaoior, tho porpotual ronowal of tho rrovaMHo?.. d.. action 



of wator in tond'mg to preserve verlicalilV, .ind llie r.-ally small 

 variation of Volooity of dillnvnl pail-. i>l' Ihc iro lowardw ill.- 

 ,-rnlro o!' a ",laciM >!' itniiMMiNo depth. I'Yoin llir?;e circnm 



ntanooH, it followN (hai i > \ fiso JH oither renewod or altogothor 

 .-\i n paird l.eforo iU vortioality IH nonnibly allootod. For tho 

 MOM Ninon, a Mick novorai foot long, innortod vortioully in tho 



