364 Alex Rosenberg
That is, they may summarize sequences of events in the lives of organisms of a
species or for that matter in organisms of higher taxa than species. Here is an
example of typical generalizations in developmental biology from Wolpert [1997,
320]:
Both leg and wing discs [inDrosophila] are divided by a compartmental
boundary that separates them into anterior and posterior developmen-
tal region. In the wing disc, a second compartment boundary between
the dorsal and ventral regions develops during the second larval instar.
When the wings form at metamorphosis, the future ventral surface
folds under the dorsal surface in the distal region to form the double
layered insect wing.Despite its singular tone, this is a general claim about all (normal) drosophila
embryos, and their leg- and wing-imaginal discs. And it is a purely descriptive
account of events in a temporal process recurring in all (normal)Drosophilalarva.
For purposes of proximate explanation of why a double layer of cells is formed in
any one particular embryo’s imaginal disc, this statement is no help. It simply
notes that this happens in them all, or that it does so ‘in order’ to eventually form
the wing, where the ‘in order to’ is implicit in the small word ‘to’.
How is the pattern of eyespot development described in the extract from Wolpert
in fact to be proximally explained? Here some of the details of a developmental
explanation may be given in order to show its special relevance to the proxi-
mate/ultimate distinction. Having identified a series of genes which control wing
development inDrosophila, biologists then discovered homologies between these
genes and genes expressed in butterfly development, and that whereas in the fruit
fly they control wing formation, in the butterfly they also control pigmentation.
The details are complex but following out a few of them shows us something impor-
tant about how proximate why-necessary explanation can cash in the promissory
notes of how-possibly explanation and in principle reduce ultimate explanations
to proximate ones.
In the fruit fly, the wing imaginal disk is first formed as a result of the expression
of the genewingless(so called because its deletion results in no wing imaginal
disk and no wing) which acts a position signal to cells directing specialization
into the wing disc-structure. Subsequently, the homeotic selector geneapterous
is switched on and produces apterous protein only in the dorsal compartment
of the imaginal disk control formation of the dorsal (top) side of the wing and
activates two genes,fringeandserratewhich form the wing margin or edge. These
effects were discovered by preventing dorsal expression ofapertous,which results
in the appearance of ventral (bottom) cells on the dorsal wing, with a margin
between them and other (nonectopic) dorsal cells. Still another gene,distal-less,
establishes the fruit fly’s wing tip. Its expression in the center of the (flat) wing
imaginal disk specifies the proximo-distal (closer to body/further from body) axis
of wing development. It is the order in which certain genes are expressed, and the
concentration of certain proteins in the ovum which explains the appearance of