The Embryo at the Wound 59
blastema fail to form when all nerves were cut, it didn't begin to form
even if a substantial number, but still a minority, remained. Also, a
salamander's leg would regrow with only motor nerves if extra motor
nerves from the belly were redirected into the stump. In addition, zoolo-
gists had found that the sensory nerve contained more fibers than the
motor nerve.
Singer counted for himself. In the thigh or upper arm, sensory fibers
outnumbered motor by four to one. The ratio was even larger at the
periphery. Then he cut them in various combinations in a long series of
experiments. Regeneration worked as long as the leg had about one
fourth to one third of its normal nerve supply, no matter in what combi-
nation. There seemed to be a threshold number of neurons (nerve cells)
needed for regrowth.
But it wasn't that simple. The limbs of Xenopus, a South American
frog unique in its ability to regenerate during adult life, had nerve fibers
numbering well under the threshold. So Singer started measuring neuron
size, and found that Xenopus had much bigger nerves than nonregenerat-
ing frogs. Another series of experiments verified the link: A critical
mass—about 30 percent—of the normal nerve tissue must be intact for
regeneration to ensue.
This finding made it pretty certain that whatever it was the nerves
delivered didn't come from their known function of transmitting infor-
mation by nerve impulses. If nerve impulses had been involved, re-
generation should have faded away gradually with greater and greater
flaws as the nerves were cut, instead of stopping abruptly when the
minimum amount no longer remained.
Singer's discovery also provided a basic explanation for the decline of
regeneration with increasing evolutionary complexity. The ratio between
body mass and total nerve tissue is about the same in most animals, but
more and more nerve became concentrated in the brain (a process called
encephalization) as animals became more complex. This diminished the
amount of nerve fiber available for stimulating regeneration in peripheral
parts, often below the critical level.
In the early 1950s, Singer applied what he'd learned to the non-
regenerating adult bullfrog. Using Locatelli's method, he dissected the
sciatic nerve out of the hind leg, leaving it attached to the spinal cord,
and directed it under the skin to the foreleg amputation stump. In two
or three weeks, blastemas had formed, and the cut legs were restored to
about the same degree as in Rose's and Polezhaev's experiments.
By
1954 Singer was ready to look for a growth-inducing chemical
that was presumed to be coming from the nerves. The most promising