40 PROF. J. ARTHUR THOMSON ON 
But what chance is there that in nature a newt should have 
its eye gouged out? To this, Weismann answers that 
newts fight furiously, at anyrate at the breeding season, and 
often injure one another; and that the larve of the large 
water-beetle (Dytiscus marginalis) often attack newts just 
behind the head. To this, I may add the observation, 
whose importance I did not unfortunately realise at the 
time, that the water-snail Planorbis may -actually kill 
salamanders, getting upon their backs and filing them with 
their radule, I do not attach importance to my addition, 
except that it seems to indicate that a fuller knowledge of 
the life of these Amphibians will show that the serious 
injury of an eye is not an unnatural casualty. 
In locusts and other insects of the same tribe, the loss 
of one of the first two pairs of legs is followed by regeneration. 
This is a well-known fact, but there is a peculiarity about it 
which will require subsequent reference. What seems more 
striking is that the posterior or third pair of legs, which are 
of great importance in jumping, are not regenerated. Now 
why should this be, that the less important may be regrown 
while the more important are not? I think the answer is 
found in the observation of Bordage that loss of the posterior 
limbs almost prevents moultings, leaves the locust exposed 
to great danger, and furthermore prevents breeding. In 
other words, the case is almost covered by my second 
addendum to Léssona’s law, “ provided that the injury be not 
fatal.” Nor can one conceive how organic provision could be 
made for an injury which prevents breeding. The pre- 
vention of breeding is a full stop in evolutionary progress. 
The last case seems to me fairly cogent ; let me now give 
one which may seem more far-fetched. Miss Helen D. 
King (Archiv Entwickelungsmechanik, vii. (1898), pp. 357— 
63. 1 pl.) notes that in the case of the starfish (Asterias 
vulgaris) an isolated arm will regenerate the whole if it has 
about a fifth of the disc left, and the not uncommon 
occurrence of “ comet-forms,” as some have called them, bears 
this out. These comet-forms consist of a fully-developed 
arm, a partially-formed disc, and four rudiments of the 
missing four arms. The investigator also notes, and this is 
the point here, that while the ventral part of an arm may 
