206 Mr. W. S. MacLeay on the Dying Struggle 



two, therefore they form two natural groups. " Omnis sectio 

 natural is," says Fries, "circulum per se clausum exhibet." 

 And as the Arachnida, on the other hand, present not two 

 circles but one, therefore they compose one natural group *. 



This is the most common shoal upon which naturalists 

 run aground. They forget the necessary and obvious in- 

 equality of chasms in different natural groups, and found their 

 own artificial groups upon this most fluctuating of all prin- 

 ciples, of which moreover they can never detect, by a ma- 

 thematical comparison of intervals, the exact value. Instead 

 of this conduct, they ought to look to the only certain prin- 

 ciple, the closure of the group by the series having returned 

 into itself. This is the only test of a natural group. From 

 the above expression of Dr. Fleming, one would really think 

 that I had never divided Arachnida, but this is not meant. 

 I certainly have not left it entire, and may possibly have di- 

 vided it just as he would; and the Doctor merely means, 

 that I have not given the divisions such dignity as he judges, 

 from certain chasms, that they deserve. My object, however, 

 was to express every natural group, and above all to demon- 

 strate the chain of continuity. All the rest, such as the com- 

 parative width of fluctuating chasms, is but leather and pru- 

 nella. 



By the by, if much more profound naturalists would at- 

 tend to the above definition of a natural group, they would 

 not so often flounder about in all the difficulties which ne- 

 cessarily attend the supposition of two determinate numbers. 

 We should not hear them agreeing with Mr. MacLeay, that in 

 most cases five is the natural number, although in some cases 

 there may be as many as seven. Were they to put their seven 

 groups each to the test of returning into itself, they would find 

 that in fact they compose only five natural ones. The deter- 

 mination of the particular number must no doubt depend on 

 observation ; but I have already proved in another place, that, 

 whatever this observed number may be, there is necessarily 

 only one. The other parts of the quinary system, such as 

 the maxim of variation, the distinction of relations of affinity 

 from those of analogy, and the progression of relations of affi- 

 nity in circles, are all truths depending upon each other, as I 

 have shown in the Linnaean Transactions. Grant one, and you 

 arrive at the other. 



* If Dr. Fleming understood what he writes about, he would have known 

 that two out of five groups always come nearer to each other than the other 

 three aberrant groups ; that Aristotle gave the name of Ptilota to the nor- 

 mal group or centrum in the present case; and that, to use the words of 

 M. Fries, " Centrum abit semper in duas series," which here are Mandibulata 

 and HaiisteUata. 



Dr 



