276 WEATHERWAX: THE ANCESTRY OF MAIZE 
the ear is unchanged, for the fact remains that Collins (p. 525, 
footnote) was using the structure of the spike of teosinte in an 
attempt to explain the appearance of ears with odd numbers of 
tows in some hybrids between maize and teosinte; and a “‘single- 
rowed”’ spike, such as that of teosinte, is of no avail in getting 
around the difficulty, for it has two rows of functional spikelets, 
not to mention the aborted ones. 
Kempton’s support (pp. 6-7) of the theory that the ear of 
maize originated by fasciation is based chiefly upon certain true- 
breeding races having bifurcated ears and a bifurcated central 
spike in the tassel. But our only evidence of the regressive nature 
of these mutants is the fasciation theory itself, and reasoning of 
this kind does not make much progress. The fact that ears hav- 
ing rows not in multiples of four could not have been formed by 
the union of spikes having four rows each is evaded by Kempton’s 
assumption (p. 7) that a row of pairs of spikelets or the pedicelled 
spikelets of two such rows were aborted. Abortion of parts isa 
common occurrence in maize, but our best evidence of this is in 
finding their rudiments; and, if abortion is thought to be respons- 
ible for the occurrence of ears with ten, fourteen, or eighteen rows, 
the burden of proof is with those who reject a more simple and 
more direct explanation of the situation. Such an explanation is 
afforded by the theory that the ear and the central spike of the 
tassel have developed from a branched inflorescence like that of 
some of the sorghums. The assumption here involved is that the 
pairs of spikelets are the remnants of reduced branches; and ma- 
terial illustrative of intermediate steps in this reduction is afforded 
by branch corn, which was unknown when Collins (p. 526) re- 
jected this theory. 
Advocates of the hypothesis that maize arose by hybridization 
find the necessity for their position in the fact that ordinary maize 
has so many primitive characteristics and seems to occupy an 
intermediate position between teosinte, which is more highly 
specialized, and pod corn, which seems to be more primitive than 
ordinary maize. 
A detailed morphological study shows that the differences in 
specialization here concerned are neither so great nor so significant 
as was supposed when the hypothesis was first advanced: but fur- 
ther, discussion of this point is waived for the present. 
