508 M. Edmond Bordage on the Tetrameric 
tated limbs in the larvee and nymphs, and I stated that very 
frequently the regenerated limbs exhibited only four joints in 
their tarsi. 
I should say that at that time I had been able to make 
but a limited number of experiments, and that I did not 
imagine that I had to deal with a general rule. Since then 
T have multiplied my observations, and I have clearly proved 
that the expression “ very frequently”? ought to be replaced 
by always. I must also make mention of the curious circum- 
stance that, in the first place, was especially instrumental in 
causing me to doubt the generality of this rule. 
In the collections of the Natural History Museum of the 
Island of Réunion (an establishment of which I am director) 
I had noticed a female Monandroptera of which the second 
pair of legs appeared themselves to be of equal length, 
although relatively shorter than the others. Moreover, the 
tarsus of one of them possessed five joints, while that of the 
other exhibited only four. My first idea was that in the 
case of these two limbs regeneration had taken place after 
autotomy. ‘This seemed then to imply that this regeneration 
preduced sometimes five tarsal joints and sometimes four. 
But measurements taken with the greatest care enabled me 
to ascertain that the limb with the pentamerous tarsus was of 
perfectly normal length, though it was owing to the length 
that, at the outset, I had been led to doubt the position given 
to the second pair of legs by the person by whom the insect 
had been set. While the anterior legs were stretched straight 
out in the direction of the axis of the body and the posterior 
ones were extended at right angles thereto, those of the 
second pair, quite doubled up, took the form of a V reversed ; 
and it was just this shortening that had made me believe that 
the dimensions were smaller. If the limb with the penta- 
merous tarsus was of normal length, that with the tetramerous 
tarsus, which seemed to be equal to it, was nevertheless 
shorter by nearly 4 millim. I was therefore led to conclude 
that the latter was the only one that formerly had undergone 
autotomy. 
To explain so trifling a difference in the length of the two 
limbs and the absolutely similar coloration that they exhibited 
one had necessarily to suppose that spontaneous amputation 
had taken place when the insect was still but a larva scarcely 
emerged from the egg. . 
In order to assure myself of the fact I collected a certain — 
number of eggs of Monandroptera tnuncans and watched 
them hatch. At birth the larvae measure about 1 centim. in — 
