Historic, archived document Do not assume content reflects current scientific knowledge, policies, or practices. _ * ae oa a - 7 ~~ | \ Am Ai 4 \s 7 a 7 —_ at oe ane -y S a), es 7 | ear Le e TH neh ee aw OP ee foe ous a a eee < Ree Sl , an 2 Ly : an , ae “- i betas, a cay ae 2 od A, = Sie Te a ao y. -aree, rye ae a ae lle 4 — In Cooperation with the Icwa Agricultural Experiment Station DEPARTMENT BULLETIN No. 1162 Washington, D. C. v September, 1923 THE ROLE OF THE GENUS RHAMNUS IN THE DISSEMINATION OF CROWN RUST. By S. M. Dietz, Assistant Pathologist, Office of Cereal Investigations, Bureau of Plant Industry. CONTENTS. Page. Page. Le ee on ie 1 | Relation of Rhamnus species to the epidemi- Relation of Rhamnus species to the epidemi- ology of crown rust—Continued. MePWEOnNCCOWSsT FUSE. 2... .=.22-5.. 2252-2056 2 Rhamuus alnifolia...... Boe eet ete Ee 15 macnamnus Cathartica .........-.-....-5.- 3 Five less important species.............. 16 Bmamnusianceolata .................... 12 ee Summary. 2 22 23222. 2e kas oo ee ee 17 iiammnus caroliniana <..-.-.....--...... 14} bAterature Cited= 22>. 9s... on. = see wee 18 x F INTRODUCTION. That certain species of Rhamnus, or buckthorn, may serve as the - alternate hosts of crown rust (Puccinia coronata Corda) is well estab- lished through the excellent contributions of De Bary (1),? Klebahn (7), Eriksson (3), and others. The réle that the alternate host may _ play in the dissemination and development of an epidemic of crown rust is not yet understood, however, because previous work had , been confined largely to laboratory cultural studies. Until recently © epidemiology studies have been impossible because of a lack of defi- nite knowledge regarding the susceptibility of our American species of Rhamnus and their geographic range in the chief oat-growing - sections of the United States. _ As early as 1894 Henning (5) observed that Rhamnus cathartica L. was dangerous to oats because of its relation to crown rust. Some _ years later, in 1912, in response to repeated recommendations that _ £#. cathartica be used as an ornamental shrub, Henning (6) stated _ that it should not be planted, citing Germany, Denmark, and Uruguay as examples of countries that have suffered epidemics of crown rust. eco wen Pape" Be y ° 1The investigations of crown rust, of which the research covered by this bulletin is a portion, are ' conducted by the Office of Cereal Investigations, Bureau of Plant Industry, United States Department of Agriculture, in cooperation with the department of botany of the Iowa Agricultural Experiment _ Station. The writer wishes to express his thanks to Dr. I. E. Melhus, plant pathologist of the Iowa j station, under whose immediate direction the investigation is conducted, for suggestions and criticisms _ during the progress of the work and the preparation of the manuscript; also to Miss Florence S. Willey, " assistant pathologist at the Iowa station, and to the Office of Cereal Investigations, for assistance in gather- ing data and in arranging the material for publication. 4 9The serial numbers (italic) in parentheses refer to ‘‘ Literature cited,’’ at the end of this bulletin. 42199°—23 2 BULLETIN 1162, U. 8. DEPARTMENT OF AGRICULTURE. Although crown rust is not of economic importance in Sweden, he | suggested that R. cathartica might cause it to become so and recom-— mended that this shrub be destroy ed. In America the literature on the role of Rhamnus in spreading crown rust has been confined to a few scattered observations. These observations, supplemented by recent cultural studies by Melhus, © Dietz, and W illey (8), emphasized the necessity for a detailed study of this problem, including such phases as the distribution of the genus Rhamnus, the prevalence of excidial infection, the dissemination of the secidiospores, and the subsequent spread of uredospores. It is AR a Ph poe ZN | Lica onal Y : We Sy Giz WME ZU i, APAAUANOS LANCEOLATA. BWRAHAMNUS in OAAAIINGS CATHARTICA. ORYIWINUS ALNIFOLIA. Fig. 1.—Sketch map of Iowa, showing the distribution of four species of Rhamnus, two native and two introduced. the purpose in this bulletin to present data dealing with these phases of the problem. RELATION OF RHAMNUS SPECIES TO THE EPIDEMIOLOGY OF CROWN RUST. The part that Rhamnus species play in starting infections of crown rust has been studied by the writer during the ‘past five years. Detailed studies have been made in Iowa throughout the season, from before any crown rust was present in the spring until the oats were harvested. These data have been substantiated by studies in the adjoining States of Minnesota, Wisconsin, Illinois, and Missouri and by periodical observations throughout the main sections of the United States in which species of Rhamnus occur. The distribution of native and introduced species of Rhamnus, or buckthorn, in Iowa is shown in Figure 1. These data are based on records obtained by the writer and his colleagues and by many other workers. RHAMNUS IN THE DISSEMINATION OF CROWN RUST. 3 -| As there is considerable difference between the various species of “| buckthorn in regard to their behavior toward crown rust, each species __ will be discussed separately. | | _ RHAMNUS CATHARTICA. - Rhamnus cathartica L. is indigenous to northern Asia and therefore is adapted to a cool climate. It is grown commonly throughout the . North Central States as an ornamental shrub or hedge plant, under such popular names as buckthorn, hartshorn, rheinberry, and way- thorn. Prior to the manufacture of barbed wire it was extensively _| planted in hedges as a substitute for fences. It was not a success as a fence, but, nevertheless, has found extensive use as an ornamental shrub. - The following places are the locations of some of the larger plant- ings of FR. cathartica in Iowa (fig. 1): Boone, Clarion, Clinton, Council Bluffs, Des Moines, Dubuque, Forest City, Harlan, Hinton, Towa City, “Marshalltown, Monticello, Muscatine, Onawa, W hiting: and in Alla- makee, Kossuth, and Pottawattamie Counties. m@ lhe ‘general use of R. cathartica through a long period of time has afforded ample opportunity for it to escape into wood lots and waste lands. As this species bears abundant fruit each year it is probable that the birds and winds have been active agents in its distribution. Noteworthy occurrences of escaped shrubs have been found in Wis- eonsin, Minnesota, South Dakota, Illinois, and Iowa, often in close proximity to oat fields. OCCURRENCE AT HINTON, IOWA. - In July, 1920, the attention of the writer was called by Dr. I. E. Melhus to a severe epidemic of crown rust at Hinton, Plymouth County, lowa. This case was conspicuous because of the extent of ‘the Rhamnus plantings and the numerous adjacent fields of oats. Northwest Iowa is fast becoming a chief oat-growing section of the “upper Mississippi Valley. Out of 2,980 acres which were surveyed in detail in this locality, 1,020 acres were sown to oats. Many varieties » sere grown, the most common of which were Albion (Iowa 103), If ichland (fowa 105), Green Russian, and Silvermine. All of these | varieties show some degree of susceptibility to crown rust. An investigation of the cause of this epidemic was made, and a buckthorn hedge 520 rods long was found. An 80-acre farm in sec- tion 18 of Lincoln Township was completely surrounded by a hedge from 10 to 20 feet high, while the adjacent section 7 in the same township had about 80 rods of hedge. By questioning neighboring -darmers it was learned that this hedge was planted about 50 years ago and that the oats grown near it had been sev erely rusted in many | menons. No definite data were obtainable for previous years, but in $920, owing to an early epidemic of crown rust, the oat crop near the | ‘a was worthless. Indeed, much of the crop was not harvested because many of the plants did not head. ~ Im 1921 a detailed study was made on the rate and distance of spread of crown rust from the infected Rhamnus shrubs in this hedge. The first pycnidia appeared on the leaves on May 13. The first idiospores were mature on May 16, and they were present in great ibundance by May 17. From one to five wcidial cluster cups were 4 BULLETIN 1162, U. S. DEPARTMENT OF AGRICULTURE. formed on each leaf, and many of the pedicels and fruits were co pletely covered with ecidia, which caused hypertrophy and subseque falling. The leaves on the upper as well as lower branches were ¥ infected. On May 22 the first uredosori appeared on the oats growing within 70 feet of this hedge. : Previous to May 22 no crown rust had been found in any other part of Jowa within 40 miles of this farm surrounded by the Rhamnus hedge. In addition, oat fields were inspected along the Lincoln Highway from central Iowa to the western border and thence north to Hinton, but no uredosori were found. The first uredospores were found at Ames, Iowa, on June 4. The percentages of rust infection were estimated on the basis of the scale in the agronomic notebook issued by the Office of Cereal Investigations of the Bureau of Plant Industry. Using one side of § the field as a base line, perpendiculars were erected every 5 rods. § Rust estimates were then made at approximately 5-rod intervals on these imaginary perpendiculars, and these percentages were recorded § at the corresponding points on the map of this area. By this method an average of one rust reading was made for each 20 square rods. Each field of oats in the vicinity was studied in this way (fig. 2). The spread of ecidiospores from this hedge during the 7-day period from May 22 to May 29 was determined by the apperance of uredosori | on the oats in adjoining fields. The possibility of increase from the J} first uredospore generation to the second was excluded, as it requires § fully seven days under average field conditions for one uredo genera- tion to develop. The infection on oats from May 22, when the first uredo infection was found near this hedge, until May 29, therefore, was caused only by ecidiospores. The hachured area in Figure 2 represents oat fields, and the intensity of the crosshatching is proportional to the degree of infec- tion. A maximum infection of 85 per cent was evident on the oats adjoining the hedge, and the infection decreased in degree as the distance from the hedge increased. Additional oat fields were inspected at increasing distances from the hedge until not a trace of crown rust was found in the fields on the final date. Where only part of the area within a field boundary is hachured oats were growO§ ing in the entire field, but no infection was found in the areas which have no crosshatching. The infection in these outer fields usually was less than 1 per cent, and no infection was found during this eee in the oat fields beyond the limits of the area represented in igure 2, which covers 5 square miles. The ecidiospores were dispersed in all directions, but chiefly toward — the southwest; that is, a heavier infection was present on the oats — southwest of the hedge, although the maximum distance to which infection was found in a southwesterly direction was but little greater than that in other directions. The maximum distance from the hedge at which ecidiospores caused infection was 452 rods, or nearly lig miles. It is probable that most of the ecidiospores were dropped — within 5 rods of the hedge, as the highest percentage of infection was — found within that distance. These data confirm an observation made — vy Eriksson (4), in which he suggests that the dispersal of spores from — an infected buckthorn is rather limited. . AXcidiospores are mucilaginous, and it is most probable that they adhere to the first surface with which they come in contact, while RHAMNUS IN THE DISSEMINATION OF CROWN RUST. 4) L 4 : vp be: edospores are dry and may be deposited and carried on a number .| of times before they are deposited finally and find favorable condi- »| tions for germination. i RRR 26-50% eeeeeceese HEDGE 4 Pee fe PASTURE BARLEY SARLEY \CLOUVER 4 a / Bx Y ‘ TIMOTHY CLOVER LLL 3 rk Fig. 2.—Sketch map of about 5 square miles surrounding the hedge of buckthorn (Rhamnus cathartica) _ hear Hinton, Iowa, showing the location of oat fields and the percentage of uredospore infection caused by the spread of zcidiospores from the hedge from May 22 to 29, 1921. (Compare with fig. 4.) FACTORS INFLUENCING THE SPREAD OF XCIDIOSPORES, fe | The degree of infection was influenced to some extent by the _ topography of the ground. In the oat field entirely surrounded by _ the hedge the grain in the ravines was more heavily infected than that on higher levels. The dew formed first in the ravines and remained there longer in the morning, which may partly explain the _ increased percentage of infection found there. =. —<—<$SS 6 BULLETIN 1162, U. S. DEPARTMENT OF AGRICULTURE. That the wind was an important factor in the spread of zecidiospore was suggested by the fact that oats growing on slopes inclined away from the hedge were not infected, while oats on slopes inclined toward the hedge were infected, even though they were a greater distance from it. As the land in the vicinity of the hedge was rolling, checks were made on the above observation, and in all cases the oats growing in the fields sloping away from the hedge were less heavily infected when a hill intervened between the hedge and the outward-facing slope. The percentage of infection was heavier in those parts of the adjoining oat fields which were most nearly on the same level as th hedge than in intervening parts of the oat fields which were lower than the hedge, even though the latter were closer to the source o infection. This, again, suggests the wind as an agent in spreading” zecidiospores. Pastures and fields of corn, alfalfa, clover, and barley (fig. 2) separated the oat fields, yet these intervening fields did not prevent the spread of ecidiospores, as oat fields 170 rods distant from the hedge were infected, even where no oat field intervened. The oats on the side of the hill nearest the hedge in all cases showed the heavier infection. A gradual decrease in infection occurred as th distance from the hedge increased, with the exception of the outward-_ facing slopes described above. The period from May 19 to 21, inclusive, during which abundant eecidiospores were formed, was marked by a southwest wind of 48 miles per hour maximum velocity (Table 1). These data show that wind may be an important factor in the greater percentage of infection in that direction. TABLE 1.—Summary of the meteorological conditions occurring in the vicinity of Hinton, Lowa, during the period from May 16 to 31, 1921. [Relative humidities at 7 a.m. and 7 p. m. were averaged.] | | | Temperature (°F.). Moisture. Wind. } | Extremes. Mean | ae _Depar-| rela- : ee r | fare | tive |: ita: Hires hour). | | Meant. |frome je u- |e | sti eet | Maxi- | Mini- | | nor- | midity | (in.) el Si /mum. | mum. | | mal. | (per Pas | Maxi- | Aver- | cent). mum. | age. 1921. | Misi on ee ae ee |. 63 42 52 9: ‘Wx 50 ~ hh arace Se 25 14.2 Ve aS A eaee \) 74 58 64| 3 78 0.02| S 27 12.7 eS SP eae Ree Se 77 58 68 6 82 26| S- 25 14.0 WE yal Oe SS Sas oe 3 BES 80 59 70 8 80 31 SW 48 | 22.1 Wie ZOGe ase Pe ee ee 87 64 76 14 (El buh “a ina Brora le SW 32 | 18.1 Mag vier 8b 58 4 2 os. | 87 69. 78 15 CAF e SW 32 21.0 [U7 27H Se ea Mei ee See 89 65 77 14 | 61 05 W 29 10.6 Maya an WRT EN 83 69 6)" 713 Bor teh tke N 12 7.49 Regie, i eee she Ftc 82 65 74| 10 79 63| NE 25 13.4 MAN 208 <2 yo bee es 83 64 (EN e, 10 76 | 68 W 25 8.0 Were 6 Ls aac Sa aa 81 62 TZ eee 77 in peeet bea) 22 6.2 LOT (i a ME seas Reena as 80 61 70 | 6 LOM de asap s N | 18 8.0 eg a STs eee ee 87 60 74] 29 ee ceeeieaeey SE | 17 8.29 IMA VEZO8 ono Bom inci sees 92 70 81 16 ASS Ns ed eee s | 29 12.2 Ma aaO: Yoo soak we Sa ae 78 61 70 3 83 ol N 21 13.0 Maiol 5s. idan: Se seesee 69 56 62 4 1 alee Se E 18 10.1 Mier; 2 ert eed 82.2 | 61.4 Tag Ga | Svea SW) {.22.:seeee The weather conditions shown in Table 1, which prevailed during the term when there was a major excidiospore infection, were RHAMNUS IN THE DISSEMINATION OF CROWN RUST. | favorable for dissemination and germination of the spores. The temperature, which averaged 8.6° F. above the normal for this | period, was favorable to the germination of the ecidiospores. The - first week in June usually marks the appearance of the zcidiospores under Iowa conditions, but their early appearance in 1921 may be due to the relatively high temperature during May. Infection was ) favored also by the high average relative humidity of 70.6 per cent sfor this period and the rainfalls on May 16, 17, 18, 19, 22, and 30, _ which were sufficient to moisten the foliage but not to wash off the | ‘spores. The heavy rains of May 24, 25, and 26 were sufficient to moisten the soil and keep a high relative humidity around the oat plants. & < THE ROLE OF GRASSES IN THE DISSEMINATION OF CROWN-RUST INFECTION. ' Earlier investigations (8) have shown that crown rust of oats is ' capable of infecting many of our native grasses in the greenhouse. It has been a common belief that the ecidiospores fell from their ' Rhamnus host to the grasses beneath and caused infection and the _ uredospores thus formed were then carried by the wind to adjoining _ oat fields. A careful study of the grasses in the entire infected area _ shown in Figure 2 was made and only one, timothy (Phlewm pratense), _ was found infected. Where growing beneath the hedge and within a Mmaximun distance of 8 rods from it this host showed a few small rust sori. It is evident that the wild grasses of northwestern Iowa _ had little or no influence on the primary spread of crown rust in the spring of 1921. The principal grass of this district is bluegrass _ (Poa pratensis), which probably is not a host for crown rust of oats. THE RATE AND DISTANCE OF UREDOSPORE DISSEMINATION. _ As mentioned earlier, the first uredosori on oats near the Rhamnus hedge appeared on May 22, and as it takes from six to seven days for a generation of uredospores to mature, the infection in the period from May 22 to 28 represents only the spread of ecidiospores. This was not a gradual spread, as all oat fields in the area of 2,980 acres covered in Figure 2 were infected in some degree by May 23. ‘This area then remained constant until May 28, after which time the first i® generation of uredospores was mature and capable of spreading infec- _ tion. As no spore traps were set, it is impossible to ascertain whether or not the ecidiospores were blown over a greater area than the one ~ in which infection was found, but it is probable that they were not _ viable when carried to a greater distance. Viable ecidiospores were _ being produced up to June 10, but as they had not been able previ- ' ously to produce infection at a greater distance than 452 rods, the more extended spread of infection must have been due to uredospores produced from excidiospores. (Table 2.) TABLE 2.—Rate of spread of crown rust to oats from a hedge of Rhamnus cathartica near Hinton, Iowa. 4 Area. ~- —— a 2 Period. _ Source of infection. | : A | Square : | Acres. | ‘miles. RIESE an Glo are Ue Se oe ee ew Sol.) 2.5) Abeidiospore.......... 2,980 5 RO te Ee ee Sk ea Sn | Uredospore.. .......... 21, 120 | 33 TAS OW IEP er eee ee ee oe Eg oe Ghee ae 80,640 | 126 memes s6 04) 915 Sis elias lle eee HT Te ed eee 407, 040 | 636 of we 8 BULLETIN 1162, U. S. DEPARTMENT OF AGRICULTURE, The six days from May 22 to 28 represent the term during which | only xcidiospores caused infection. The remaining 13 days shown in — Table 2 are the period in which the rapid spread was due to succes- sive uredospore generations. The direction and distance of infection in the period from June 5— to 7 are shown in Figure 3. The lack of infection southwest of Le Mars apparently is due to the fact that no oats were grown for several J square miles. This area also is heavily wooded, and it may be pos- — sible that the trees offered some obstruction to the spread of the, uredospores. In a detailed survey made of the area shown in Figure 3 no other specimens of Rhamnus were present than those in the hedge already described. : Eas eS Py | | SPREAD FRO/A L ow ge RX QQ | BUCKTHORN HEDGE KW N \N | y NE Pee ) |_| CLOUER WN Ws WG re | |AISTURE, | CORN mes SS WS } | i AN 85%, DOT RUST 30 ORT RUST 95qp OAT RUST 100% CHT RUST | ecscesssess LLDGE SNS ~ \ SS | ALFALFA SS SN \N CORN SS SS SS \ \ aS SN SN WAL samen 7 37 7 rad TURE PASTURE NS: FASTUR. Ss Ss eS be “> Af, SS 2: 2 SS eaten S é YY f 4A f tf, YY Nf hf, (J, 4 4 x > S&S XS ORY RR LL, S&S = 255 CORN | CORN TEERCORN ; i= Stated ie eee | Mose eee SSE 3 a0 3 CKO ! BSI nines YX | Uitte: j RICKI RIK, CORN | CORN | PRSTURE. iSSSeuss: | } KCK? XL SS OK KOS ‘\ SISOS SNS FAST URE = Fs WX \\' ie con CORN \ \ \ CORN \\ | a oRN| (CORN CORN a jae CORN KX | | STORE| CORN a SSS See Fic. 4.—Sketch map of about 5 square miles surrounding the hedge of buckthorn (Rhamnus cathartica) near Hinton, lowa, showing the location of the oat fields and the percentage of infection which had occurred to the end of June7. (Compare with fig. 2.) around the roots. Unless this soil is removed, the plant will continue to live and put out leaves, and these may bear heavy excidial infec- tion. However, the shrubs pulled in this way seldom survive until the second spring after being removed. 42199-—23-——2 10 BULLETIN 1162, U. S. DEPARTMENT OF AGRICULTURE, OTHER EPIDEMICS. A 10-acre pasture at Clarion, Iowa, in which Rhamnus cathartica has escaped from cultivation, has been under observation for the past three years, although it was not studied in as much detail as the planting at Hinton. The cultivated shrubs were 10 to 20 feet high, and the escaped shrubs ranged from mere seedlings to a height of 4 feet. During this 3-year period both the cultivated and escaped shrubs have borne ecidial infection, although only a maximum infection of 10 per cent occurred on any of the shrubs. In the spring of 1921 a more detailed study was made of this area. The ecidial stage appeared on the Rhamnus leaves previous to the advent of the uredospore stage on*oats. The oats adjoining the hedge showed the first crown rust infection in this locality. The oats in the portion of the field farthest removed from the hedge developed the least infection. All evidence Fic. 5.—Hedge of Rhamnus cathartica near Hinton, lowa. The oats adjoining this hedge had 100 per cent of crown-rust infection. The crop was barely knee high, many plants not heading. The yield from this field did not pay the expense of harvesting. Photographed July 1, 1921. pointed to the fact that R. cathartica acting as an alternate host caused the initial infection of crown rust on these oats. In August, 1919, and in July, 1920, and July, 1921, several oat fields were inspected in the vicinity of Janesville, Avalon, Emerald Grove, and Delavan Lake, Wis., where many shrubs of R. cathartica have escaped from cultivation. Hedges were found along the roadsides, and seedlings were scattered throughout the vicinity. During the springs of 1919 and 1921 the ecidial infection was so severe that the shrubs were entirely defoliated. In 1919 the oats in the fields adjoining the buckthorns were not worth cutting. The oats inspected within a 2-mile radius were severely injured. In 1920 the xcidial stage was again present on the bushes, and this infection was disseminated to oats. The next year, 1921, witnessed another severe loss in the yield of oats which was traceable to infection from the neighboring buckthorn shrubs. . 4 : RHAMNUS IN THE DISSEMINATION OF CROWN RUST. EE _ MMeidiospores have been collected on FR. cathartica in the vicinity of | Madison, Wis., during the past four years, and each year the oats ' adjoining the hedges have been heavily infected. ‘ |. The role played by R. cathartica was demonstrated also at Warren, | Minn., by Dr. Henry S. Conard, of Grinnell, Iowa, in July, 1919. A _ few weeks later further investigations were made at Warren, where 15 bs ra : I 3 / 5 hs : a Fic. 6.—T wo samples of oats from different parts of the same field. 2 The right-hand sample, showing 100 per cent infection with crown - rust, was cut adjoining the hedge shown in figures 5and7. The 5 left-hand sample came from the distant side of the same field. The = intervening oats showed a gradual decrease in percentage of infection and increase in height and size of panicle as the distance from the hedge increased. Photographed July 1, 1921. hedges of R. cathartica were studied. The field of oats nearest to a hedge was about half a mile distant. The rust infection in this field was 90 per cent. [very field of oats inspected within a radius of 10 miles was heavily infected with crown rust. Teleutospores were developing on the oats nearest the hedges, while the rust on the oats farther away was only in the uredospore stage. The county agent | : 12 BULLETIN 1162, U. S. DEPARTMENT OF AGRICULTURE. | | at Warren reported that the owners of these hedges had sent him many samples of buckthorn leaves heavily infected with the ecidial stage of crown rust during the spring of 1919. Additional observations have been made at Fargo, N. Dak., Noiieche 3 H and Freeport, Ill., and Duluth, Minn., where Ff. cathartica has caused — i local epidemics of crown rust. F a | RHAMNUS LANCEOLATAe { i Although unknown in the Old World, Rhamnus lanceolata Pursh is © i ‘perhaps the most widely distributed native species. It is commonly — known as lance-leaved buckthorn and grows most commonly along ~ | narrow streams and shaded dry runs. Its range extends from eastern 4 5 | 4 4 d 4 4 Yas he Maid ner SE Ada eae Fic. 7.—Plants of Rhamnus cathartica from the hedge shown in figure 5. The great lumps of dirt left on the roots by the stump puller enabled the plants to remain alive. They produced leaves in the spring, and these had abundant ecidial infection. Photographed July 1, 1921. Nebraska and through central Iowa eastward to Pennsylvania and southward to Florida. This shrub has been collected from the following places in Iowa (see fig. 1): Adel, Bellevue, Bloomfield, Centerville, Council Bluffs, Dubuque, Elkader, Indianola, Le Claire, Monmouth, Morning Sun, Mount Pleasant, Osceola, Oskaloosa, Ottumwa, Postville, “Salem, Sioux City, Volga, Winterset; also at “The Ledges’”’ in Boone County, in Decatur County, in the north side of Marion County, in Shelby County, and along the South River in Warren County. It is not frequently found adjoining oat fields. a STUDIES IN IOWA. As R. lanceolata is one of the most widely distributed native species it is important to know to what extent it bears the ecidial stage of - initial infection on oats and other grasses. _ throughout the region in which this shrub grows. _ studies were carried on near Indianola, as this town is located about RHAMNUS IN THE DISSEMINATION OF CROWN RUST. 13 crown rust and whether or not it is instrumental in starting the The situation has been studied in detail-in Iowa, and many observations have been made The principal in the center of the area in which R. lanceolata occurs in Iowa. These studies were further substantiated by observations at Bloom- field and Osceola. Each spring since 1917 field observations have been made on R. lanceolata in the vicinity of Indianola to ascertain its relation to a yearly infection of crown rust on oats. Mcidial infection was found on these shrubs every year. However, the maximum infection, in 1919, showed only 10 per cent of the leaves bearing ecidia. When the berries were infected the ecidial cups covered the entire fruit and often the pedicel. These shrubs, which varied from mere seedlings Re RSE ° 0 a°,cao S, QO XS BOW °, °, 0% \ O SX O oJ <5 go OK 505 S05 S © ° , 50% SRS ‘ot SOIR KS 5 Fig. 8.—Sketch map of an oat field and adjacent shrubs of Rhamnus lanceolata near Indianola, Iowa, showing the spread of uredospores from ecidiospores, causing an infection of 15 per cent on the oats nearest the shrubs, with gradually diminishing infection to less than 1 per cent at the distant side of the field. to trees 15 feet high, bore ecidia on the topmost leaves as well as on those near the ground. - Observations also were made in a locality 34 miles southwest of Indianola, where, along the banks of a small stream, hundreds of seedlings as well as mature plants were found. ‘This stream flows along a 90-acre field which was sown to oats in 1919. Shrubs bear- ing ecidia grew within 100 feet of this field. An infection of 15 per cent occurred on the oats nearest the shrubs, with a gradually dimin- ishing infection at greater distances until only a trace or less than 1 per cent was found in the most distant portions of the field. (Fig. 8.) This field had a southeastern exposure toward the bordering stream. The area showing the heaviest infection also was most favored by heavy dews, which fall earlier in the evening and remain longer in the morning, thus providing more favorable conditions for infection / 14 BULLETIN 1162, U. S. DEPARTMENT OF AGRICULTURE. to develop in the area nearest the buckthorns. Numerous grasses | bordered this area, though Phlewm pratense L. and Calamagrostis canadensis (Michx.) Beauv. were the only two grasses which were infected with crown rust. The flora directly beneath the Rhamnus shrubs consisted largely of sedges. At still another location, about 7 miles removed from this field Phleum pratense, Calamagrostis canadensis, Sphenopolis (Katonia) obtusata (Michx.) Scribn., Agrostis hyemalis (Walt.) 8: 5. (2) aim Avena sativa L. were infected with crown rust when rowing near Rhamnus lanceolata. In addition, Festuca elatior L. was found infected with crown rust in 1921. Although many cases have been inspected during the past five years, that at Indianola in 1919 is the only one which gave positive evidence that &. lanceolata is instru- mental in spreading the crown rust to oats under field conditions. Although ecidial infection occurs on this species at Osceola and at Bloomfield, Iowa, each year, the situation at the latter place is dif- ferent from those previously discussed. At Bloomfield, which is in the extreme southern portion of the State, H. H. Plagge found uredospore infection on oats in 1919 previous to the appearance of ecidial infection on I. lanceolata. It is probable that crown rust here had hibernated as mycelium or in the uredospore stage, a rare occurrence in Iowa. R. lanceolata is not used as a hedge or ornamental shrub and there- fore is seldom found growing in the immediate vicinity of oat fields. It should be noted also that the area comprising the habitat of this shrub lies principally outside the main oat-growing section of the Middle West. STUDIES IN ARKANSAS AND MISSOURI. From April 2 to 10, 1919, an extensive field study was made in the vicinities of Fort Smith, Prescott, and Little Rock, Ark., to deter- mine whether this shrub was instrumental in starting the annual infection of crown rust in that section. At this time R. lanceolata was just coming into leaf and no infection was evident, although uredosori were common on winter oats. As crown rust often lives over winter in Arkansas, this shrub seems to play only a minor rdle in the initial appearance and subsequent dissemination of crown rust there. Acidial infection occurs on R. lanceolata in Missouri annually, as shown by studies at Arlington. It was not found infected at Moun- tain Grove, in the southern part of the State, during the springs of 1919 and 1920. Here, as in Iowa, the shrub seldom + erows hear oat fields, and hence its opportunity for aiding the spread of crown rust is not extensive. RHAMN JS CAROLINIANA. Rhamnus caroliniana Walter is a native species extending eastward from Texas, Kansas, and Missouri, the latter State being the northern limit of its range. It is abundant on shaded hillsides and stony ridges of eastern ‘Kentucky and Tennessee. It most closely resembles R. lanceolata, which it largely supplants in the southeastern section of the United States. The hills surrounding Knoxville, Tenn., bear an abundant growth of this shrub, varying in height from 6 inches to 30 feet or more. - RHAMNUS IN THE DISSEMINATION OF CROWN RUST. 15 Observations made in April, 1919, just as these shrubs weré coming into leaf, showed no evidence of infection. Oats growing in an adjoining field showed no uredospore infection. In July, however, this field was heavily infected with crown rust, although there still was no trace of ecidial infection on the Rhamnus. Carl Kurtzweil reported excidial infection on Rhamnus at this location in 1920. Observations were made at Lexington, Ky., and in the Missouri Botanical Garden at St. Louis in 1919, but no ecidial infection was ‘present. Dr. Harrison Garman, in a personal conversation, said that R. caroliniana was heavily infected at Lexington, Ky., and. Arthur reported finding zcidia on /?. carolinana in a garden at La Fayette, Ind., in 1905. Although repeated search has been made, the writer has never found ecidia on this shrub in the field. Plantings on the campus at Ames, Iowa, have never developed xcidiospores, though oats heavily infected with crown rust grew near the Rhamnus. Greenhouse experiments in which teleutospores from Avena sativa were used as inoculum gave only pycnidia (8). However, in the spring of 1922 a few cluster cups were formed on Rf. caroliniana in the greenhouse trials with the teleutospores from Avena sativa. Because of the limited investigations made, the réle of R. carolin- dana in spreading crown rust of oats is thus far not well understood. The data available suggest that 1t is not an important alternate host for the specialized form of crown rust occurring on the cultivated oat. RHAMNUS ALNIFOLIA. Rhamnus alnifolia L’ Her., or dwarf alder, which is largely confined to swamps and shaded banks of streams, is primarily a northern species, reaching the southern limits of its distribution in the northern portions of Iowa, Nebraska, and Illinois. In lowa it has been collected at Decorah and along the Yellow River in the north- eastern part of the State. It grows abundantly also in Door county, in northeastern Wisconsin. About 25 oat fields were inspected in that county, but crown-rust infection was almost entirely absent. A maximum infection of 10 per cent was found, but this was confined to a few small local patches. In the lowlands surrounding Sturgeon Bay, Wis., PR. alnifolia grows in great profusion. The grass flora of this locality consists mostly of Calamagrostis canadensis, which occurs beneath these shrubs and often extends to the edge of the lowlands where oats are grown. This locality has been under observation for the past four years to deter- mine the possible relationship between the ecidial infection on this buckthorn and the uredospore infection on oats. Abundant teleuto- spore and uredospore material on Calamagrostis has been collected there in every year except 1920, when the infection was light. On May 24, 1919, Calamagrostis canadensis infected with uredo- spores of crown rust was growing in the vicinity of FR. alnifolia at Sturgeon Bay, Wis. The Rhamnus at this time was just coming into leaf and showed no traces of infection. Greenhouse tests (8) have shown that crown rust on C. canadensis is only occasionally able to infect oats. Although it was evident that «ecidia on R. alni- folia were directly responsible for infection on Calamagrostis, this buckthorn probably does not act as a direct agent in the distribution of crown rust to oats. 16 BULLETIN 1162, U. S. DEPARTMENT OF AGRICULTURE. FIVE LESS IMPORTANT SPECIES. Various factors and combinations of factorscause Rhamnus frangula, R. smithu, R. purshiana, R. californica, and R. crocea to be less important in the dissemination of crown rust to oats than are the species previously mentioned. These species of Rhamnus have a limited distribution and their habitats lie principally outside of the oat-producing sections. They frequently are found at high altitudes, far removed from cultivated fields, which precludes the possibility of their becoming infected under ordinary field conditions. RHAMNUS FRANGULA, Rhamnus frangula L., commonly known as black alder, berry alder, and arrowwood, is a native of Europe. It often is used in the United States as a hedge and ornamental plant and has escaped commonly in bogs and wet places on Long Island, in northern New Jersey, and in Ontario, Canada, according to Britton (2). It is not common in the Middle West, but has been reported by J. B. Mayer as escaped from cultivation at Hull, Iowa, which is the only place known to the writer where Ff. frangula occurs as an escape in Iowa. According to greenhouse trials (8) R. frangula developed pycnidial infection, but never ecidial infection, when exposed to teleutospores from species of Avena and Calamagrostis, two of the most widely distributed crown-rust hosts in the oat-producing region. The escaped plants near Hull in the northwestern part of the State and cultivated plants at the Missouri Botanical Garden in St. Louis both have been inspected, but there was no ecidial infection at either place. Although but limited observations have been made, it seems probable that A. frangula is of little importance in the dissemination of crown rust in the Middle West. RHAMNOUS SMITHII. Rhamnus smithii Greene, a thick-stemmed dwarf shrub bearing lanceolate leaves, is confined to a few thousand acres near Pagosa Springs in southwestern Colorado, where it grows very abundantly. It is rarely found near cultivated fields. It is cultivated in Boulder, Colo., where it thrives and often reaches a height of 12 feet. At Pagosa Springs, Colo., a relatively high percentage of ecidial infection was present in 1920 and 1921. Teleutospores of crown rust were collected at that place on Agro- pyron smithii, Koeleria cristata, Bromus ciliatus, and Trisetum sp. by Miss Florence Willey in July, 1921. Shrubs of R. smith growing near these grasses were heavily infected. Later in the season uredo- spores were collected on Bromus sp. growing beneath heavily infected bushes. Although the relationship of this alternate host to these grass species must be verified by controlled greenhouse study, it is probable that it is not of much importance in spreading crown rust to oats, as little of this cereal is grown in the vicinity. Its response to infection with teleutospores from oats under greenhouse conditions (8) suggests its close relationship to R. lanceolata. It may be that R. smithii is only a form of R. lanceolata which has become adapted to the high altitudes of the Rocky Mountain region. ‘ RHAMNUS IN THE DISSEMINATION OF CROWN RUST. i RHAMNUS PURSHIANA. Rhamnus purshiana DC. is a western species occurring commonly in California, Oregon, Washington, Idaho, and Montana. Its natural habitat is at an elevation where cultivated fields do not occur. Although R. purshiana has been infected with the ecidial stage of crown rust in the greenhouse, it probably is not a source of infection in oats, as it grows at high altitudes in the mountains while oats are grown in the valleys. RHAMNUS CALIFORNICA. This species and the variety tomentella are largely confined to the western coast of the United States. The field observations on Rham- nus californica Esch. are limited, but probably it plays little or no part in the spread of crown rust to oats. It grows in a warm tem- perate climate where the uredospore stage of crown rust exists throughout the entire year. RHAMNUS CROCEA. Experiments (8) have shown that Rhamnus crocea Nutt. will produce zecidiospores when exposed to infection with teleutospores obtained from Calamagrostis canadensis. The results of experiments conducted in 1922 gave a like degree of infection with teleutospores from Festuca elatior and Avena satiwwa. These data indicate that this species might be able to play an important réle in the spread of crown rust to oats. However, #. crocea is limited to a small area in southern California and to certain islands off the coast and is unimportant because it commonly does not occur within an oat-producing area. SUMMARY. Rhamnus cathartica L., a native of Asia, has long been used as an ornamental shrub in this country and has escaped commonly into wood lots and waste lands throughout the north-central and north- eastern United States. This shrub commonly bears ecidia of crown rust (Puccinia coronata Corda), and was found to be the cause of severe epidemics of crown- rust in Iowa, Wisconsin, Minnesota, and Illinois. The epidemic at Hinton, Iowa, caused by ecidial infection on a 520-rod hedge, was so severe in 1921 that the oats in adjoining fields were not harvested. ‘Topography, wind, and other climatic conditions during the time the zcidiospores are forming probably are the most important factors in the spread of crown rust. Wild or cultivated grasses growing near the infected hedge apparently had little influence on the spread of the rust. Aicidiospores from R. cathartica caused infection to a distance of 452 rods, or nearly 14 miles, in seven days, or before the first uredo- spores were ripe. The infection was heavy near the hedge and progressively lighter as the distance from it increased. After the first generation of uredospores was mature, the spread of infection was veryrapid. Thirteen days later oats were infected at a maximum distance of 53 miles from the hedge. 18 BULLETIN 1162, U. S. DEPARTMENT OF AGRICULTURE. Rhamnus lanceolata Pursh is perhaps the most widely distributed native species. It occurs commonly from eastern Nebraska across south-central Iowa to Pennsylvania and southward to Florida. It is not used as an ornamental shrub and seldom grows near cultivated fields. Though it bears wecidia every year, the case at Indianola, Iowa, is the only one known to the writer where this species was responsible for the initial infection of crown rust of oats. ecidiospores collected on Ff. lanceolata were capable of infecting Phleum pratense, Calamagrostis canadensis, Agrostis hyemalis, and Festuca elatior under field conditions. As uredospores overwinter in the South, it is probable that this species plays little part in the initial infection by crown rust there. Rhamnus carolimana is another native species, extending eastward from Texas, Kansas, and Missouri to the Atlantic coast. Apparently it does not bear ecidiospores every year, but in recent greenhouse experiments cluster cups were produced when it was exposed to infection by teleutospores from oats. RR. alnifolia is a native northern species bearing heavy ecidiospore infection of crown rust where Calamagrostis canadensis grows near it. Although it has been found to aid the spread of crown rust to this grass, no case is known where it was a factor in spreading the specialized form of crown rust which occurs on oats. Five other species of buckthorn, one introduced and four native, namely, PR. frangula, R. smithu, R. purshiana, R. californica, and RF. crocea, are relatively unimportant in the dissemination of crown rust to oats. R. frangula has escaped from cultivation in the north- eastern United States. The others are species of very limited distribution in the far West and usually distant from fields of oats. R. smithii and R. purshiana bear ecidia under field conditions, and grasses near the former species become infected. LITERATURE CITED. (1) Bary, A. DE. 1866. Neue Untersuchungen iiber die Uredineen . . . Jn Monatsber. K. Preuss. Akad. Wiss. Berlin, 1865, p. 15-49, 1 pl. Biblio- graphical footnotes. (2) Brirron, NATHANIEL Lorp. 1907. Manual of the flora of the Northern Statesand Canada. Ed. 3. xxiv, 1122 p. New York. (3) Ertksson, JAKOB. - 1894. Ueber die Specialisirung des Parasitismus bei den Getreiderostpil- zen. In Ber. Deut. Bot. Gesell., Bd. 12, p. 292-331. Laitteratur- verzeichniss, p. 330-331. (4) 1904. Ueber das Vegetative Leben der Getreiderostpilze. I. K. Svenska Vet.-Akad. Handl., Bd. 37, no. 6, 19 p., 3 pl. (5) HENNING, Ernst. 1894. Nagra ord om olika predisposition fér rost 4 sid. In K. Landtbr. Akad. Handl. och Tidskr., Arg. 33, p. 205-217. (6) 1921. Rhamnus cathartica b6r icke odlas. (Rhamnus cathartica ought not to be cultivated). Jn Svensk. Land. No. 5. (7) KurBann, H. 1892. Kulturversuche mit heterécischen Uredineen. Jn Ztschr. Pflanzen- krank., Bd. 2, p. 332-343. Biblographical footnotes. (8) Meruyus, I. E., Dierz, 8.M., and WiLLEy, FLORENCE. 1922. The alternate hosts and biologic specialization of crown-rust in Amer- ica. Jowa Agr. Exp. Sta. Research Bul. 72, p. 211-236, 2 fig. Bibliography, p. 235-236. ORGANIZATION OF THE UNITED STATES DEPARTMENT OF AGRICULTURE. Assistant Secretary Bemecerciary of Agriculture.........-..-..------- Henry C, WALLACE. C. W. PUGSLEY. Director of Scientific Work ................+-. E. D. Batt. Director of Regulatory Work........-.-.-.---- merector of Eatension. Work .....-..-.-.----- Forest Service ...-. Weather Bureau... Bureau of Agricultural Economics ........--.- Henry C, Taytor, oe Bureau of Animal Industry .......-.----- _.--- JOHN R. Mouter, Chief. BBureau of Plant Industry ................-.-- Witiram A. Taytor, Chief. w-ee eee ee eee ew ee ee ee ee te ew ee ee tt He ee ee Geaeea i Marvin, Chief. W. B. GREELEY, Chief. IEA OREMISHTIY = 2522-52 Soe ee Water G. CAMBELL, Acting Chief. Bureau of Soils... mepureau of Entomology......---.------------- L. O. Howarp, Chief. Bureau of Biological Survey........-.-------- K. W. NEtson, Chief. Bureau OS ETT Sg a THomas H. MacDonatp, Chief ‘Bureau of Home Economics........-.---.---- LOUISE STANLEY, Acting Chief. Fined Nitrogen Research ee erate. Se Seep eet F. G. Corrre.y, Director. Division of Accounts and Disbursements... ..-- A. ZAPPONE, Chief. OS ee ee eee Mitton Wuitney, Chief. CLARIBEL R. Barnett, Librarian. ‘Federal Horticultural Boord........-. Ae. Be C. L. Maruarr, Chairman. | | | rh Insecticide and Fungicide Board ...........--- J. K. Haywoop, Chairman, Packers and Stockyards Administration .....-.-- CHESTER Morritt, Assistant to the Grain Future Trading Act Administration.... - Secretary. | Be SOM =: 8 2s ee eS R. W. Wiis, Solicitor. This bulletin is a contribution from B Bureau eerteetIuSt NH 02 22) a 2s 2 + WiuiramM A. Taytor, Chief. Office of Cereal Investigations.........-.-- CARLETON R. Batt, Cerealist in Charge. 19 ADDITIONAL COPIES OF THIS PUBLICATION MAY BE PROCURED FROM THE SUPERINTENDENT OF DOCUMENTS GOVERNMENT PRINTING OFFICE WASHINGTON, D. C. AT 5 CENTS PER COPY PURCHASER AGREES NOT TO RESELL OR DISTRIBUTE THIS COPY FOR PROFIT.—PUB. RES. 57, APPROVED MAY 11, 1922 ¥ WASHINGTON : GOVERNMENT PRINTING OFFICE ; 1923