pathogen and the fact that, when present and 
effective, it may obscure an underlying dan- 
gerous, supersusceptibility to any new race 
that arises to attack it. 
Horizontal resistance is typified by the 
stable, polygenic resistance to late blight of 
potato. The one great virtue of horizontal 
resistance is its permanence. Its liability is 
its usually multigenic inheritance, making it 
of rare occurrence and costly to recognize 
and recover from hybrid progeny. 
CURRENT APPROACHES TO 
BETTER USE OF VERTICAL 
RESISTANCE 
One attempted approach to more successful 
use of monogenic, vertical resistance in the 
wheat rusts has been the transfer of genes 
from related grass genera to the genus 
Triticum. Some of us have held hope that 
resistance derived from distant wheat rela- 
tives might be more enduring than that pro-= 
vided by genes found occurring naturally in 
wheat. Such a resistance from Aegilops um= 
bellulata was supplied by Sears (28) in the 
Transfer variety. No races of leaf rust (P. 
recondita) attacking this variety have been 
found in nature in North America. Samborski 
(27), however, produced such a virulent clone 
by selfing a Canadian culture of this species 
on the alternate host, and recently the Rocke- 
feller Foundation (24) reported Transfer to be 
susceptible in the field in Argentina. Thus, 
this Aegilops resistance also proves to be 
undependable. Several wheat derivatives of 
Agropyron elongatum, including Agrus and 
Agatha with an immune reaction to leaf rust, 
are being intensively studied in interstate and 
international tests. Thus far, they have re- 
mained resistant. There is, however, no firm 
ground for hope that vertical resistance de- 
rived from interspecific and intergeneric 
crosses will provide the needed dependability 
of resistance. Knott and I-sun Shen (17) have 
recommended that continued attempts be made 
to provide new sources of resistance to stem 
rust, by hybridization of wheat with related 
species and genera, to supplement the rela- 
tively few vertical resistance genes now known, 
The breeding of "multiline" varieties, pos- 
sessing vertical resistance, has been proposed 
by Jensen (15) and Borlaug (1) as an approach 
122 
to more stable rust resistance. Here, several 
vertical resistance genes are individually 
added to single, near isogenic lines that are 
then composited for commercial use, as the 
incidence of new races may require at anyone 
time. This approach is based mainly on the 
theory that no one race will develop rapidly 
enough on susceptible plants surrounded by 
resistant plants to cause severe damage. The 
practical proof of the theory of disease control 
by multiline varieties largely remains to be 
provided. The method is laborious and would 
require most of the resources of most breeding 
programs and thus limit much that might 
otherwise be done in creating inherently better 
crop varieties. The method is highly conserva- 
tive, restricting the breeder to older crop 
genotypes of proven performance. The mixture 
of resistance types might provide the ideal 
situation for breeding new races by vegetative 
recombination of virulence genes of the old 
rust races, if mechanisms of vegetative re- 
combination operate under field conditions, as 
proposed by Nelson et al. (18), and Ellingboe 
(9). The Rockefeller Foundation (24) has an- 
nounced the release of the first multiline 
wheat variety, Miramar 63, with resistance to 
both stem and stripe rust in Colombia, South 
America. The degree of success and perma= 
nence attained in controlling the development 
of rust epidemics with this first multiline 
variety will provide valuable guidance to 
future breeders. 
At Purdue we have used a mature-plant 
vertical resistance of a polygenic type to 
produce effective mature-plant resistance in 
the soft red winter wheat varieties Knox and 
Vermillion, grown on about 2 million acres in 
the Southern and Eastern United States during 
most of the past decade. Their resistance, 
derived from Chinese C.I. 6223, has been 
shown by Unraw et al. (33) to be conditioned 
by genes on four chromosomes. Our studies 
have also indicated polygenic inheritance in 
both Chinese and Knox with segregatants in 
Fo hybrid populations ranging continuously 
from 0 to 100 percent field infection, During 
1962 to 1964, races have appeared that cause 
up to 30 percent infection on Knox and Vermil- 
lion as compared with 100 percent on suscepti- 
ble checks, This resistance, despite its poly- 
genic inheritance, is therefore of the 
hypersensitivity and vertical type. Thus, 
