Studies in the Calcite Group. 239 



this change is not a uniform one. It might be considered that 

 the structure and molecular volume of normal dolomite can be 

 retained without chang-e until slightly over 10 per cent of the 

 isomorphous molecules are present, while after that point it must 

 yield to their influence and show a progressive change. Under 

 these conditions the line giving the change in molecular volumes 

 or specific gravity with change in composition would be a curve 

 and not a straight line. The broken line in Fig. 4 represents 

 such a possible curve in the case of dolomite. As far as the 

 writer is aware no careful survey has ever been made of the ends 

 of a series of isomorphous salts that would serve to establish or 

 contradict this possibility. The magnesite-siderite and siderite- 

 rhodochrosite series discussed above were examined to see if 

 they gave any evidence along this line. Only a few of these 

 analyses showed a percentage of their chief constituent of 90 per 

 cent or over and further the differences between the molecular 

 volumes of the end members of the two series are about one half 

 as great as in the case of the dolomite series. What evidence 

 there was, seemed to confirm the above theory that the lattice 

 remained constant for the end sections of the series, but it was 

 not strong enough to enable one to make any definite statements 

 in regard to it. Indeed, for the reasons given above, it could not 

 be expected that the evidence in these series would be at all clear. 

 The straight solid line on Fig. 4 shows the variation in the 

 molecular volumes of dolomite when it is considered that this 

 change is uniform for the entire series. The theoretical specific 

 gravities for the dolomites have been calculated from their anal- 

 yses under this assumption and are given in column IV of Table 

 V and the differences between these values and those measured 

 are given in column V of the same table. It is interesting to 

 compare these differences between the measured and theoretical 

 values for the first twenty-five analyses with those derived under 

 the assumption that the molecular volume remains uniform for 

 this portion of the series. The average difference between the 

 measured and calculated values for these twenty-five analyses 

 as given in column V is .022 or if their positive and negative 

 characters are considered, +.016. It will be seen that while the 

 average error in this case is within allowable limits, it is notably 

 greater than that given for the other case in a previous 

 paragraph. 



