660 MR. J. LEWIS BONUOTE U.N THE [-^P^'- ^^j 



Miifl ITT. showing' tiMC(S of tlie other ;i pices, does not invnlidate the 

 reasonhig as groups (ii.) and (iii.) are intennediate forms. 



ISlus rattus. 



Cumparisoii of Indimi and Egyptian Forms. 



'11 le next question to arise is, what is the significance of tlie 

 third and smallest apex at 29 and 30 ram. On reference to my 

 ))apei" qnoted above in which I had divided J/, rattnn into three 

 suhgronps, we find the hind-foot measurements as follows : — 



Jalorensis subgroup, hill form. Hind foot 30 nnu. (29-31 '5). 

 Ji'afescens ,, tree form. „ 33 mm. 



Griseiventer „ house form. ,, 35 mm. 



These measurements, it will be noted, coi-respond exactly to 

 the three apices in the ctu-ves of the Egyptian rats; so thiit we 

 are, I think, justified in considering the small apex at 29 to 

 represent the hill form (Jalorensis sid:>group) of the Oriejifcil 

 Kegion, but which has become, in the level country of Egypt, 

 unrecognisable externally. 



This is a point which is of extreme importance from a deeper 

 and more general point of view, for here, in a low-lying country, 

 A\'e have traces of a variety which in another part of the woi-ld 

 has become the predominant form in the hills. Noav, if con- 

 tinuous variation Avere the only method of evolution, we should 

 find that in a level country all variations suitable to the hills 

 Avoidd lie eliminated, so that, in coui'se of years, no trace of them 

 would remain as they would be swamped by the more suitable 

 forms, but apparently this has not liappened in the present 

 instance. The small-footed variety is still present in Egypt, 

 jiermeating the race, and still existing in sufticiently strong 

 numbers to spring up again (juickly and become a dominant form 

 slinnld it ever find itself under suitable conditions. 



Although, perhaps, not absolutely conclusive, this aftbrds con- 

 siderable evidence that the three subgroups of ;!/. raitns, which 

 1 fiist difl'erentiated in 1903 and which are strongly confirmed 

 ]iy the piesent work, piobal)ly arose as mutations. 



'inhere is yet a, fnithei- point. One of the chai-acters of the 

 Jdlorriisls subgroup (the hill t'onu) is the pure white hairs of the 

 underparts. In this case it is associated with a small hind foot, 

 but in Egy|)t we find it in J/, lecloram, which has the laigest 

 hind foot of the three forms. It is therefore evident that the 

 colour of the underparts is a charactei' which has not. always 

 tieen inherited in connection with the size of the foot, but may 

 occur in any one of the suV)groups, so that in one country it may 

 be characlieristic of the hill bum. in another of the tree form, and 



* I ;nn awavc tliat on tlir rvidPticP sn far hroupht forvvavrl it mip'ht br arpupd 

 that, it thr colnur of the iindri parts is- not al«a.vs found in correlation with thp 

 >amp si/cd hind foot, it is a > haiacter rf bnt little use in distingiii>hin? thr puh- 



