Equation 13 holds for those factors (except the incubation period) which effect r but not the gradient. With such a range of factors
F. Vulnerability and Resistance of the Host to Disease;
a Matter of Natural Selection
The question of the type of plant in which epidemics are most likely to be common can be probed from another angle: the geographical and ecological position of the plant before it was taken into cultivation. This has been much better understood in forestry than in agriculture.
The fundamental concept is that of vulnerability. A state of vulner-ability means that circumstances are favorable for attack and harm by appropriate pathogens. Vulnerability must not be confused with susceptibility. Indeed, in nature they are to a great extent mutually ex-clusive: a species can be in a very vulnerable state or it can be very susceptible, but it is unlikely for long to be both vulnerable and sus-ceptible. Natural selection will see to that.*
To give an example of geographical vulnerability, North American plants have been vulnerable to North American pathogens, and, if brought into cultivation in their natural habitat, are likely to have some measure of resistance to them. Extended, the argument brings one to the well-known fact that epidemics are more common if either the host or the pathogen has been introduced.
To give examples of ecological vulnerability, ecologically dominant plants—plants in a more or less pure stand—are more vulnerable than those in mixed stands because inoculum can more easily spread between them. Perennial plants are more vulnerable to systemic disease than an-nuals because systemic disease (usually) persists for the rest of the life of the plant.
Dominant long-lived trees—redwoods, pines, eucalypts, and the like—are, because of their ecological state, highly vulnerable to viruses and on the geological evidence many have been so continuously for
mil-* Natural selection is implicit in the whole argument. This is the centenary of the publication by Darwin and Wallace of their essay, and plant pathologists can have little satisfaction in the thought that 100 years have gone by without the greatest concept in biology having left more than a scratch on their subject.
lions of years. That they have survived destruction by harmful viruses implies great resistance, and up until now no virus disease has been recorded in them. The argument for dominant long-lived trees holds also, although sometimes to a lesser extent, for dominant perennials of any sort, such as the dominant grasses of veld and prairie. Great vul
nerability can be expected to have required great resistance, which seems to accord with the facts, because there does not seem to be a single record of natural infection (let alon^ an epidemic) by a harmful virus of any plant species that has achieved ecological dominance and is growing in its natural environment. Exceptions may yet be found, but they are unlikely to alter the general conclusion that there have been epidemics of plant virus diseases only because we have taken into cultivation a lot of annuals and other plants with no background of ecological dominance in nature.
Tolerance must be grouped with resistance here, and the discovery of harmless latent viruses in ecologically dominant perennials would not invalidate the argument in any way.
These examples illustrate the connection between vulnerability and resistance. One point is still very obscure. Geographical vulnerability can explain why epidemics are more common with foreign pathogens or in foreign hosts. But can ecological vulnerability sometimes lead to general resistance, even against foreign pathogens? So far, e.g., the red
woods, pines, and eucalypts have remained free from harmful virus diseases despite international commerce and movement of viruses. It would be a mistake to reject the possibility of general resistance as far-fetched just because one cannot for the moment point to the mechanism.
A P P E N D I X : A SUGGESTION F O R T H E CO NTROL O F S T E M R U S T O F W H E A T
When this chapter was written, it was realized that stem rust of wheat, caused by Puccinia graminis tritici, behaved in many ways like blight of potatoes. Rowell (1957), testing a method of inoculating wheat with rust spores carried in oil, had published an S-shaped progress curve for an attack of race 15B on Marquis wheat. From this curve one cal
culates that r was about 50% per day between July 4 and 10, 1956. But the work of Hay den (1956) had not been seen. Without knowledge of r and the appropriate equations Hayden was not in a position to see fully the capital significance of his results.
When Race 15B reached epiphytotic levels, it was observed that the varieties Lee and Sentry were less severely damaged than Marquis, Mida, Carleton, and Nugget, even though all these varieties (Lee and Sentry included) were fully "susceptible," in the sense that the
infec-Hon types with Race 15B were 3 or X + to 4. Hayden inoculated these varieties with Race 15B when all had headed except Marquis and Carle-ton, which were in the boot stage. Relatively little difference in yield resulted between plots inoculated with heavy and light concentrations of spores with all varieties except Lee and Sentry. When exposed to light concentrations of spores, the yield of these two varieties was de
pressed much less than that of all other varieties. As inoculum was in
creased from light to heavy, the yield of Lee and Sentry more closely approached that of the four other varieties. (By equation 11 the effects of varying amounts of inoculum at the start are greatest when r is least.) Hayden inoculated plants with Race 15B in the center of plots. The rate of spread of rust from these centers was lowest in Lee among the bread wheats and in Sentry among the durums. (By equations 8 and 13, with a given incubation period and environmental conditions, the rate of spread of a given pathogen is determined by r.) Hayden inoculated plants with spores. On the "susceptible" varieties except Carleton, which was not included in the test, the inoculations produced fewer lesions per culm on Lee among the bread wheats and on Sentry among the durums.
From these results there is little doubt that Lee and Sentry have lower r values, and that these relatively low r values are at least in part the result of a resistance to infection which cannot possibly be reflected in the accepted criterion of "susceptibility": an X, 3 or 4 type of reaction. The resistance of Lee and Sentry to Race 15B is com
parable with that of the potato variety Noordeling to Phytophthora infestans, discussed in Section ΙΙΙ,Ε. In each case there is a certain ap
parent tolerance in the sense that the varieties fare better than others in an epidemic, and this tolerance is in reality partial resistance mani
fested as a low value of r. Partial resistance has long been recognized and is considered to be polygenic and not greatly affected by the race of the fungus. It will be assumed that this is so.
There is nothing original in a suggestion that tolerance in some form or other might profitably be used to lessen losses from rust. The suggestion has often been mooted. But our recommendation is not just tolerance, valuable though it be, but cumulative delay. A rust epidemic adequately delayed is no epidemic at all.
Let us rewrite equation 5 as follows:
. , 2 3 0 . x 2(l - Xl )
The time taken for an epidemic in a field to develop from any given level of infection xt to any other given level x2 is inversely proportional
to r. But the same idea applies to an epidemic in its whole course from field to field, from country to country. If the values of r of all "sus
ceptible" varieties could be halved, then every step which before took 1 day would now take 2, other things being equal. One pictures the whole epidemic process occurring step by step in the same order as before, but with each step now taking twice the time. An epidemic which before needed 2 months to travel from Mexico to Canada would now need 4, other things being equal. Epidemics starting from infected barberries or wild grasses would be delayed in the same way.
Let us consider the errors. The reasoning is not true of the relatively short time that spores are actually air-borne between fields. Another error comes from lesions' growing old and sterile. This error is on the safe side, i.e., it makes equation 5 undervalue the benefit of reducing r. The condition that other things be equal is unattainable. Here too the error is likely to be on the safe side. Rust epiphytotics occur as a result of particularly favorable combinations of climatic conditions, and these are less likely to extend over, say, 4 months than over 2. Finally, an error comes from the incubation period's effect on the disease progress curve (Section II, A , 2 ) . Here too the error is on the safe side, because if, e.g., factors counterbalancing this effect make the curve for a high-r variety sigmoid, they will make the curve for a low-r variety progressively flatter than sigmoid.
Because the delay is cumulative with the progress of the epidemic, it might seem that on the main North American rust track the benefits of delay would increase from south to north: that Canada would gain more than Mexico. Perhaps this is true. But the south in turn gets in
oculum from the north, so the process is not just one way.
Suppose it were agreed that after some specified date no new vari
eties would be released unless they had relatively low r values. There would be an immediate local benefit to farmers who planted these varieties, a benefit of the same sort as that gained by those who planted Lee and Sentry during the recent epidemics of Race 15B. But hopes would go far beyond that—to the time when low r varieties had ac
cumulated along the rust track and delayed general epidemics out of existence. If all states along the rust track pursued a policy of en
lightened local self-interest and released only low r varieties, they could expect both an interim and a final dividend. That is our sug
gestion.
It would probably be easiest to determine r comparatively. Lee or Sentry might, for example, be used as a standard for testing against Race 15B those varieties that have a "susceptible" reaction to this race. The easiest, but not the only, way of producing suitable new varieties would
be to use breeding material "susceptible" to one or other race. But it is not the aim of this appendix to discuss methods of breeding; it is to describe the benefits that could be expected from stable polygenic resistance manifested as relatively low values of r and to state how this resistance could best be used.
R E F E R E N C E S
Anonymous. 1949. W. African Cacao Research Inst. Tafo, Gold Coast, Ann. Rept.
1 9 4 7 / 8 : 2 1 .
Anonymous. 1957. Weed em and reap. Monsanto Mag. 3 7 : 29-33.
Beaumont, Α., J. H. Bant, and I. F. Storey. 1953. Potato spraying trials in York
shire, 1947-n51. PUnt Pathol. 2 : 56-60.
Bonde, R., and E. S. Schultz. 1943. Potato cull piles as a source of late-blight infection. Am. Potato J. 2 0 : 112-118.
Broadbent, L. 1957. Investigations of virus diseases of Brassica crops. Cambridge Univ. Press, London and New York. 94 pp.
Chester, K. S. 1943. The decisive influence of late winter weather on wheat leaf rust epiphytotics. Phnt Disease Reptr. Suppl. 1 4 3 : 133-144.
Chester, K. S. 1946. "The Cereal Rusts." Chronica Botanica, Waltham, Massa
chusetts. 269 pp.
Chester, K. S. 1950. Plant disease losses: their appraisal and interpretation. Phnt Disease Reptr. Suppl 1 9 3 : 190-362.
Christ, R. A. 1957. Control plots in experiments with fungicides. Commonwealth Phytopathol. News 3 : 62.
Cochran, W. G. 1936. The statistical analysis of field counts of diseased plants.
Suppl. J. Roy. Statist. Soc. 3 : 49-67.
Doncaster, J. P., and P. H. Gregory. 1948. The spread of virus diseases in the potato crop. (Brit.) Agr. Research Council Rept. Ser. No. 7 : 189 pp.
Freeman, G. H. 1953. Spread of diseases in a rectangular plantation with vacancies.
Biometrica 4 0 : 287-296.
Gaumann, E. 1946. "Pflanzliche Infektionslehre." Verlag Birkhauser, Basel. 611 pp.
Grainger, J. 1950. Forecasting outbreaks of potato blight in West Scotland. Brit.
Mycol. Soc. Trans. 3 3 : 82-91.
Grainger, J. 1956. Host nutrition and attack by fungal parasites. Phytopathology 4 6 : 445-456.
Gregory, P. H. 1945. The dispersal of air-borne spores. Brit. Mycol. Soc. Trans.
2 8 : 26-72.
Gregory, P. H. 1948. The multiple-infection transformation. Ann. Appl. Biol. 3 5 : 412-417.
Hansen, H. P. 1950. Investigations of virus yellows in beets in Denmark. Trans.
Danish Acad. Tech. Sci. 1950: 68 pp. (Abstr. in Rev. Appl. Mycol. 3 0 : 399).
Hayden, Ε. B. 1956. Progressive development of infection by Puccinia graminis var.
tritici Eriks. and E. Henn. (Guyot) on certain varieties of wheat and the rela
tion of stem rust to yield. Univ. Minnesota Graduate School Thesis. 103 pp.
Heald, F. D., and R. A. Studhalter. 1914. Birds as the carriers of the chestnut blight fungus. /. Agr. Research 2 : 405-422.
Hewitt, W. Β., N. W. Frazier, J. H. Freitag, and A. J. Winkler. 1949. Pierce's dis
ease investigations. Hilgardia 1 9 : 207-263.
Hildebrand, Ε. M. 1953. Yellow-red or X-disease of peach. Cornell Univ. Agr.
Expt. Sta. Mem. 3 2 3 : 54 pp.
Hildebrand, Ε. M., G. H. Berkeley, and D. Cation. 1942. Handbook of virus diseases of stone fruits in North America. Mich. Agr. Expt. Sta. Misc. Publ. 76 pp.
Hull, R. 1952. Control of virus yellows in sugar beet seed crops. /. Roy. Agr. Soc.
Engl. 1 1 3 : 86-102.
Large, E. C. 1945. Field trials of copper fungicides for the control of potato blight.
I. Foliage protection and yield. Ann. Appl. Biol. 3 2 : 319-329.
Large, E. C. 1952. The interpretation of progress curves for potato blight and other plant diseases. Phnt Pathol. 1 : 109-117.
Large, E. C , R. E. Taylor, I. F. Storey, and A. H. Yule. 1954. Spraying trials in the potato-growing area around the Wash, 1948-53. Phnt Pathol. 3 : 40-48.
Leach, J. G. 1940. "Insect Transmission of Plant Diseases." McGraw-Hill, New York. 615 pp.
Leach, L. D. 1938. Determining the sclerotial population of Sclerotium rolfsii by soil analysis and predicting losses of sugar beets on the basis of these analyses.
J. Agr. Research 5 6 : 619-631.
Limasset, P. 1939. Recherches sur le Phytophthora infestans (Mont.) de Bary.
Ann. epiphyt. 5 : 21-39.
Liming, Ο. Ν., E. G. Rex, and K. Layton. 1951. Effects of a source of heavy infec
tion on the development of Dutch elm disease in a community. Phytopathology 4 1 : 146-151.
Linford, Μ. B. 1943. Influence of plant populations upon the incidence of pine
apple yellow spot. Phytopathohgy 3 3 : 408-410.
McCallan, S. E. A. 1946. Outstanding diseases of agricultural crops and uses of fungicides in the United States Contribs. Boyce Thompson Inst. 1 4 : 105-115.
McClean, A. P. D., and J. E. van der Plank. 1955. The role of seedling yellows and stem pitting in tristeza disease of citrus. Phytopathohgy 4 5 : 222-224.
McClintock, J. Α., L. O. Kunkel, and Η. H. Thornberry. 1951. Peach rosette. In Virus diseases and other disorders with virus-like symptoms of stone fruits in
North America. U. S. Dept. Agr. Agr. Handbook 1 0 : 7-10.
Mills, W. R. 1940. Phytophthora infestans on tomato. Phytopathology 3 0 : 830-839.
Moore, P. W., E. Nauer, and W. Yendol. 1957. California scaly bark disease of citrus. Calif. Agr. 1 1 : 8-9.
Moore, W. C. 1943. The measurement of plant diseases in the field. Brit. Mycol.
Soc. Trans. 2 6 : 28-35.
Newhall, A. G. 1938. The spread of onion mildew by wind-borne conidia of Peronospora destructor. Phytopathohgy 2 8 : 257-269.
Norman, G. G. 1956. Where do we stand on budwood certification? Citrus Ind. 3 7 : 16-17.
Peace, T. R. 1957. Approach and perspective in forest pathology. Forestry 3 0 : 47-56.
Posnette, A. F. 1947. Virus diseases of cacao in West Africa. Ann. Appl. Biol 3 4 : 388-402.
Posnette, A. F. 1951. Virus research at the West African Cacao Research Institute, Tafo, Gold Coast. Tropical Agr. London 2 8 : 133-142.
Posnette, A. F. 1953. Virus diseases of cacao in West Africa: the present position.
Rept. 13th Intern. Hort. Congr., London, 1952 pp. 1224-1230.
Riker, A. J. 1951. The spread of oak wilt in local areas. Phytopathohgy 4 1 : 30.
Rowell, J. B. 1957. Oil inoculation of wheat with spores of Puccinia graminis var.
tritici. Phytopathohgy 4 7 : 689^-690.
Sprague, R. 1953. Powdery mildew on apples. Plant Diseases, U. S. Dept. Agr., Year
book Agr. pp. 667-670.
Stevens, Ν. E. 1939. Disease damage and pollination types in "grains." Science 8 9 : 339-340.
Stevens, Ν. E. 1948. Disease damage in clonal and self-pollinated crops. /. Am. Soc.
Agron. 4 0 : 841-844.
Storey, I. F., and A. E. Godwin. 1953. Cauliflower mosaic in Yorkshire 1950-51.
Phnt Pathol. 2 : 98-101.
Thomas, Η. E., C. E. Scott, Ε. E. Wilson and J. H. Freitag. 1944. Dissemination of a peach mosaic. Phytopathology 3 4 : 658-661.
Todd, H. 1940. A note on random associations in a square point lattice. /. Roy.
Statist. Soc. 7 : 78-82.
van der Plank, J. E. 1946. A method for estimating the number of random groups of adjacent diseased plants in a homogeneous field. Trans. Roy. Soc. S. Africa 3 1 : 269-278.
van der Plank, J. E. 1947. The relation between the size of plant and the spread of systemic diseases. Ann. Appl. Biol. 3 4 : 376-387.
van der Plank, J. E. 1948. The relation between the size of fields and the spread of plant disease into them. I. Crowd diseases. Empire J. Exptl. Agr. 1 6 : 134-142.
van der Plank, J. E. 1949a. Some suggestions on the history of potato virus X. /.
Linnean Soc. London, Botany 5 3 : 251-262.
van der Plank, J. E. 1949b. The relation between the size of fields and the spread of plant diseases into them. Empire J. Exptl. Agr. 1 2 : 18-22.
van der Plank, J. E. 1958. Some epidemiological consequences of systemic infec
tion. Plant Pathol: Problems and Progress 1908-1958.
van der Plank, J. E., and Ε. E. Anderssen, 1945. Kromnek disease of tobacco: a mathematical solution to a problem of disease. Union S. Africa Dept. Agr.
and Forestry Sci. Bull 2 4 0 : 6 pp.
van der Zaag, D. E. 1956. Overwintering en epidemiologie van Phytophthora in
festans, tevens enige nieuwe bestrijdingsmogelijkheden. Tijdschr. Phntenziek-ten 6 2 : 89-156.
Waggoner, P. E. 1952. Distribution of potato late blight around inoculum sources.
Phytopathology 4 2 : 323-328.
Wellman, F. L. 1937. Control of southern celery mosaic in Florida by removing weeds that serve as sources of mosaic infection. U. S. Dept. Agr. Tech. Bull 5 4 8 : 16 pp.
Wolf, F. A, 1935. "Tobacco Diseases and Decays." Duke Univ. Press, 454 pp.
Zentmyer, G. Α., P. P. Wallace, and J. G. Horsfall. 1944. Distance as a dosage factor in the spread of Dutch elm disease. Phytopathology 3 4 : 1025-1033.