PATTERNS IN THE INTEGUMENT OF INSECTS
81
10. An Interpretation of Gradient Phenomena
It is assumed that tracheae and surface integument are alike with respect to the properties under discussion. Any hypothesis to account for
the properties described as a gradient must take into account (1) that
the gradient has local stability, it is not a property continually and solely
imposed by a segment or the animal as a whole, (2) wound cells may lose
their gradient properties, (3) wound cells gain the gradient of their unwounded neighbors, (4) in isolated grafts which rearrange their pattern
without being influenced by the host, the gradient may be reoriented with
its highest level in the center of the graft, (5) the intersegmental membranes or nodes are the source of some factor influencing the extent of
growth of other regions, and (6) this factor operates in one direction in
a way not accounted for by diffusion alone. There are probably several
related phenomena involved.
First, we may suppose that there is some morphological correlate to
the orientation, something to orient the gradient, to cause a difference in
permeability in one direction. We may expect that cell adhesions will be
involved. It is probably significant that migrating cells which have lost
their adhesions to one another, may also have lost their gradient properties. The reimposition of the surrounding gradient upon wound cells
which have restored the normal density could then be in part an infectious spread of junctions with appropriate distribution. A morphological counterpart to the orientation would also help to explain the local
stability of the gradient.
Second, to obtain a gradient something must be maintained at a higher
level in the anterior cells. The local stability of the gradient means that
this process probably takes place in each cell. There are two possibilities.
Either each cell makes something which survives locally in an equilibrium
in proportion to its level in the gradient, or all cells make about the same
amount which is then transported preferentially in one direction because
of a polarized property of the transporting mechanism working against
the gradient. The latter hypothesis describes the organizing properties
of the nodes and the polarity of tracheae, but is inadequate to explain the
continued isolation of grafts transposed in the axis. We may be wrong in
making too close a parallel between the intersegmental membranes and
the nodes. The intersegmental membranes separate regions with similar
polarity but different W e i in the gradient, while the nodes are junctions
between tracheae of opposite polarity. Both hypotheses may therefore be
true, and we should probably separate the phenomena.
On the other hand, the experiments on tracheae do not exclude the
possibility that the polarity is a result of a gradient. If there is a growth
stimulating factor with a polarized distribution arising at the interseg-
81
10. An Interpretation of Gradient Phenomena
It is assumed that tracheae and surface integument are alike with respect to the properties under discussion. Any hypothesis to account for
the properties described as a gradient must take into account (1) that
the gradient has local stability, it is not a property continually and solely
imposed by a segment or the animal as a whole, (2) wound cells may lose
their gradient properties, (3) wound cells gain the gradient of their unwounded neighbors, (4) in isolated grafts which rearrange their pattern
without being influenced by the host, the gradient may be reoriented with
its highest level in the center of the graft, (5) the intersegmental membranes or nodes are the source of some factor influencing the extent of
growth of other regions, and (6) this factor operates in one direction in
a way not accounted for by diffusion alone. There are probably several
related phenomena involved.
First, we may suppose that there is some morphological correlate to
the orientation, something to orient the gradient, to cause a difference in
permeability in one direction. We may expect that cell adhesions will be
involved. It is probably significant that migrating cells which have lost
their adhesions to one another, may also have lost their gradient properties. The reimposition of the surrounding gradient upon wound cells
which have restored the normal density could then be in part an infectious spread of junctions with appropriate distribution. A morphological counterpart to the orientation would also help to explain the local
stability of the gradient.
Second, to obtain a gradient something must be maintained at a higher
level in the anterior cells. The local stability of the gradient means that
this process probably takes place in each cell. There are two possibilities.
Either each cell makes something which survives locally in an equilibrium
in proportion to its level in the gradient, or all cells make about the same
amount which is then transported preferentially in one direction because
of a polarized property of the transporting mechanism working against
the gradient. The latter hypothesis describes the organizing properties
of the nodes and the polarity of tracheae, but is inadequate to explain the
continued isolation of grafts transposed in the axis. We may be wrong in
making too close a parallel between the intersegmental membranes and
the nodes. The intersegmental membranes separate regions with similar
polarity but different W e i in the gradient, while the nodes are junctions
between tracheae of opposite polarity. Both hypotheses may therefore be
true, and we should probably separate the phenomena.
On the other hand, the experiments on tracheae do not exclude the
possibility that the polarity is a result of a gradient. If there is a growth
stimulating factor with a polarized distribution arising at the interseg-