362 Professor H. H. Swinnerton— 
Considering next a group of organisms which includes repre- 
sentatives of two or more lineages, only one case seems conceivable, 
viz. the one in which the rate of change is different for the several 
series. If the different series were so intimately correlated, that no 
change could take place in one without involving all the others in 
an equal amount of change, the members of one line of descent would 
be indistinguishable from those of another. As a matter of fact, 
even when there is striking homceomorphy, the members of the 
different lineages can still be sorted out from one another. This 
is because one lineage differs from another in the rate of change in 
one or more homologous structural units. In this case the graphs, 
representing several individual organisms, will not develop out of 
one another ; on the contrary, they will cross one another (Fig. 2, I). 
Ambulacruim. Interambulaceurn, 
Fic. 3.—Graphic representation of some of the differences existing between 
several genera of Paleozoic echinoids. The Roman numerals I-VI stand 
for the drawings having the same numbers in Fig. 1. The Arabic number 
14 in the left column stands for drawing number VII. 
EXAMPLES OF GRAPHICAL REPRESENTATION. 
To illustrate the application of the methods just described to the 
graphical representation of fossils, four examples will now be taken. 
For the first the facts already drawn from the Paleozoic echinoids 
will be used. In this case the facts will be taken as they stand without ' 
reducing them to percentages of possible change. The other three 
will be taken from Carboniferous corals. In these the facts will 
consist mainly of measurements, and will be expressed in the form 
of percentages. 
