524 CRITICAL rERIODS IK THE HISTORY OF THE EARTH. 



Laws and forces are indeed uniform, but phenomena are nearly always par- 

 oxysmal. The forces of volcanoes and earthquakes, of lightning and temp- 

 est, are uniform, but the phenomena are paroxysmal. Winds at the earth's 

 surface, where the resistance is great, blow in puffs. A thin sheet of water 

 over a smooth sloping surface runs in waves. The law may be illustrated 

 a thousand ways. In all cases where an accumulating force is opposed by 

 a constant resistance, we have phenomena in paroxysms. 



But whatever be the cause, the/ac^of paroxysmal movement of organic 

 ^volution is undoubted. All along the course of geological history, from 

 beginning to end, even when the times were quietest, where the record is 

 fullest and apparently without any missing leaf, species come and go and 

 others take their place, and yet only rarely do we find any transition steps. 

 If this were merely once or twice or thrice, or to any extent exceptional, it 

 might be explained by loss of record here and there, but it occurs thousands 

 and tens of thousands of times. JSIow, if evolution moves only at uniform 

 rate, if it takes one hundred thousand years to transmute one species into 

 another (as it certainly does when evolution is moving at its usual rate), if 

 there are at least one hundred thousands stejDs (represented each, of course, 

 by a whole generation of many individuals) between every two consecu- 

 tive species, it is simply incredible that all the individuals representing the 

 intermediate steps, so infinitel}^ more numerous than the species they con- 

 nect, should be so generally, almost universal!}^, lost. But the phenomena, 

 as we find them, are easilj^ understood if a few generations represent the 

 transition step, and many generations the permanent form. 



A similar rapid, almost sudden, appearance and extinction of genera, 

 families, and higher groups at certain horizons, are also common. In these 

 cases the intermediate steps of transition are often found, and constitute, 

 in fact, the chief demonstrative evidence of the truth of evolution. But 

 the difficulty on the assumption of a uniform rate of evolution is none the 

 less here, for the time required to evolve a new genus or a new family is, of 

 course, immensely greater than in the case of a new species. 



We will illustrate the difficulties of the ordinary vi-ew by one striking 

 example. In the Upper Silurian, in the midst of a conformable series, — 

 where if there bo any break, any lost record, surelj^ it must be very small, 

 — appear suddenly, without premonition, fishes; not a connecting link be- 

 tween fishes and any form of invertebrates, but perfect, unraistakable 

 fishes. Here we have, therefore, the appearance not only of a new class, 

 but of a new sub-kingdom or type of structure, Yertthrata. Now, to change 

 from any previously existing form of invertebrate, whether worm, crusta- 

 cean, or mollusk, into a vertebrate, by a series of imperceptible steps rep- 

 resented by successive generations, — steps so imperceptible that it would 

 take one hundred thousand of them to advance from one intermediate spe- 

 cies to an other, — would require an amount of time which is inconceivable 

 to the human mind, aiid a number of steps, each be it remembered, repre- 

 sented by thousands of individuals, which can scarcely be exi)reesed by fig- 



