Oct., 1875.] ELLIOTT SOCIETY. 89 



axy sub-groups might be named from their leading members, and we would 

 have for line —3, the Phosphorus and Vanadium sub-groupp, for line —2, the 

 Sulphur and Chromium sub-groups, for line + 1, the Kalium and Sodium sub- 

 groups, for line +2, the Calcium and Magnesium sub-groups. As the mem- 

 bers of each pair of sub-groups are intermingled in the corresponding line of 

 the Table, if it were desired to separate them when each Une or Group was un- 

 der special consideration the secondary group might be written in inverted 

 order, thus : 



for Une -2, S=32, Se=79, Te=128, Os=199, W=184, Mo=96, Or=52.5, 

 for line +2, Ca=40,Sr=87.5, Ba=137, Pb=207, Cd= 112, Zn=65, Mg=24, 



Gl=:9.3, 



the atomic weights first increasing successively, culminating in the member 

 belonging to Series K, then diminishing through the members of the secondary 

 group. In the two sub-groups, Cr, Mo, W, and V, Cb, Ta, it is to be remarked 

 that the difference between the second and third members of each group, 88, 

 is almost exactly double the difference between the first and second members. 

 Each of the negative members of series B, namely F, O, N, and C, though 

 closely allied to the primary sub-group of their respective lines, is yet so dis- 

 tinctly marked by several characteristics that they may be regarded as forming 

 in those lines each a third sub-group (consisting of a single member,) lor 

 which it may be convenient to adopt some distinctive epithet. In the Iron 

 Group we may also distinguish two sub-groups, Cr. Mn, Fe in one, Co, Ni, 

 Cu in the other ; some writers would prefer to interchange Fe and Co. The 

 first of these groups Cr, Mn, Fe, is marked by the power of forming b Jth bases 

 and acids ; by their basic character they are allied to the other members of the 

 same group, and by their acid character they are allied through Cr to S and the 

 Sulphur group, and so an alliance is indicated between the upper and lower 

 portions of the Table, which have been previously shewn in contrast. For 

 these reasons the symbol Cr and its atomic weight is inserted in both portions, 

 and if in Series D, Cr were replaced by Mn or Fe, the atomic weight so intro- 

 duced would in some respects accord better with the results of the formulae 

 given above. It may be objected here that the relation between S and Cr used 

 as the ground of alliance between the groups, is exhibited by ihem not in their 

 dyad but in their hexad character, and then they ought to be placed in line —6 

 as well as — 2 ; so also the relations which induced us to place V in the same 

 group with P, hold between them in their pentad character, and that therefore 

 both ought to be written on line — 5 to express this relationship. The objection 

 is, of course, well founded, and will apply to many other cases than these ; and 

 can only be met by maintaining the position already taken in the early part of 

 this paper ; that to obtain the advantages presented by a single table, we must 

 assent to the sacrifice of some of those gained by the use of several tables ; the 

 inverse is also true, that if we desire the advantages given by several tables, we 

 must sacrifice those belonging to a single table, and these are in the present 

 VOL. n. 12 rublished April, 1886. 



