It is likely that feathers evolved from a conical shaped tubercle rather th
an a plate-like structure. Although the morphology of the presumably most p
rimitive feather is unknown, minimal conditions for its production include
the cellular capacity to synthesize feather proteins (=phi -keratin) which
provides the molecular phenotype, and a follicular mechanism for production
and assembly of molecular and gross structure, Once the mininal structural
element, presumably recognizable as a barb, existed, a variety of phenotyp
es followed rapidly. A tubercular growth center of appropriate size could p
roduce a simple barb-like element, with cortex and medulla, This might be r
ecognized externally as a bristle, but need never existed as a separate mor
phological unit. Rather, if individual placodes gave rise to multiple barb
ridges that fused proximally, a structure resembling natal down would have
resulted, Subsequent differentiation is controlled by the follicular symmet
ry, and the feather shape is regulated by barb length. Barb length is direc
tly related to growth period. As feathers appear to grow at roughly similar
, size independent rates, shape is determined by individual barb growth per
iods. The simple fusion of individual proto-barbs would produce a morpholog
y identifiable as natal down. Although this might be the simplest feather s
tructure, others could emerge quickly, perhaps simultaneously, a consequenc
e of the same redundant processing. Once the machinery existed, broad pheno
typic plasticity was possible. I constructed a feather phylogram based on t
hese conditions, the fossil record, and ontogeny. I organized the subsequen
t changes in morphology by perceived complexity. The changes are simply ind
ividual responses to similar processes that might be time (when in ontogeny
) and space (where on body) dependent.