7.12.
AUTORADIOGRAPHY
251
cannot at the present time be expected even to approach the great resolution possible with the electron microscope. Autoradiographic resolution cannot be regarded as ever better than the size of the silver halide
crystals in the original emulsion, and may be degraded further if the
developed silver grains exceed the size of the crystals. However, the distribution of developed grains can be studied statistically so that probabilities of error in identifications of loci often can be reduced effectively
by such means. Caro (1962) defends the thesis that an autoradiographic
resolution of approximately 0.1 u. is about all that can be expected under
the best of circumstances with presently available commercial emulsions.
However, a resolution of 0.5 u. is easily obtainable. Unfortunately, even
this would not be realizable with many atoms other than tritium. Tritium
emits (3-particles with a particularly low energy (0.02 Mev). This minimizes the amount of their scatter, and ionization effects are maximal
very close to the emitting source. Thus it is this atom in particular which
can affect silver halide crystals in a monolayer emulsion on top of an
ultrathin section. If one tried to use high energy particles they would
tend to pass through the very thin photographic emulsion without producing an ionizing effect, or they sometimes would produce ionization
relatively far removed laterally from the source. Thus it is that of all
the common radio-atoms, tritium is by far the most useful, and it is
fortunate indeed that it can be so easily incorporated in many molecules
of biological interest. Carbon-14 and sulphur-35 also have promise, but
their (3-particle energies of 0.16 and 0.17 Mev, respectively, exceed that of
tritium by a factor of ten.
Another factor that enters into the application of autoradiographic
techniques to electron microscopy is the limitation of tissue mass in an
ultrathin section, and therefore the amount of radioactivity that can be
realized practically. If exposure times are to be at all reasonable, the
tissue must be radiologically "hot" indeed. This is all the more so since
the photographic emulsion also must be thin. It behooves the investigator,
then, to use as thick sections as he can afford to. This puts a premium
on a microscope having an 80 or 100 kv power supply. It also indicates
that it will be a radiographic advantage to sacrifice some electron microscopic quality, and use methacrylate sections rather than cross-linked
embedments (although this writer has indeed seen some excellent autoradiographs of material embedded in the latter). It is common practice
to select light gold (0.1 u) sections for these purposes. Even so, Hay and
Revel (1963) found they needed a 3 week exposure time to obtain rich
AUTORADIOGRAPHY
251
cannot at the present time be expected even to approach the great resolution possible with the electron microscope. Autoradiographic resolution cannot be regarded as ever better than the size of the silver halide
crystals in the original emulsion, and may be degraded further if the
developed silver grains exceed the size of the crystals. However, the distribution of developed grains can be studied statistically so that probabilities of error in identifications of loci often can be reduced effectively
by such means. Caro (1962) defends the thesis that an autoradiographic
resolution of approximately 0.1 u. is about all that can be expected under
the best of circumstances with presently available commercial emulsions.
However, a resolution of 0.5 u. is easily obtainable. Unfortunately, even
this would not be realizable with many atoms other than tritium. Tritium
emits (3-particles with a particularly low energy (0.02 Mev). This minimizes the amount of their scatter, and ionization effects are maximal
very close to the emitting source. Thus it is this atom in particular which
can affect silver halide crystals in a monolayer emulsion on top of an
ultrathin section. If one tried to use high energy particles they would
tend to pass through the very thin photographic emulsion without producing an ionizing effect, or they sometimes would produce ionization
relatively far removed laterally from the source. Thus it is that of all
the common radio-atoms, tritium is by far the most useful, and it is
fortunate indeed that it can be so easily incorporated in many molecules
of biological interest. Carbon-14 and sulphur-35 also have promise, but
their (3-particle energies of 0.16 and 0.17 Mev, respectively, exceed that of
tritium by a factor of ten.
Another factor that enters into the application of autoradiographic
techniques to electron microscopy is the limitation of tissue mass in an
ultrathin section, and therefore the amount of radioactivity that can be
realized practically. If exposure times are to be at all reasonable, the
tissue must be radiologically "hot" indeed. This is all the more so since
the photographic emulsion also must be thin. It behooves the investigator,
then, to use as thick sections as he can afford to. This puts a premium
on a microscope having an 80 or 100 kv power supply. It also indicates
that it will be a radiographic advantage to sacrifice some electron microscopic quality, and use methacrylate sections rather than cross-linked
embedments (although this writer has indeed seen some excellent autoradiographs of material embedded in the latter). It is common practice
to select light gold (0.1 u) sections for these purposes. Even so, Hay and
Revel (1963) found they needed a 3 week exposure time to obtain rich