Journal of Heredity, 398-403 (1915)
Work at Dominion Experimental Farms Begun by Late Dr. William Saunders—
Mostly with Apples—Many Hardy Types Produced—Work with Vegetables and Ornamentals

Dominion Horticulturist, Central Experimental Farm, Ottawa, Canada.

THE breeding of horticultural plants at the Dominion Experimental Farms was begun when the late Dr. Wm. Saunders brought from London, Ont., in 1888, a large collection of bush fruits and grapes which he had accumulated as a result of his work in cross-breeding begun in 1868. Since 1888 a continuous effort has been made to originate new varieties of fruits, vegetables and flowers which would be more useful in some parts of Canada than anything available from other sources. Canada had up to that time depended almost entirely for new varieties of fruits on foreign countries and while this is true to a large extent today, much has been done by the Dominion Experimental Farms to develop new plants. While the main purpose has been to obtain new varieties of commercial value, the possible discovery of underlying principles has been kept constantly in mind.

As the main work in breeding has been with the apple, the greater part of this article will be devoted to giving an account of what has been done with this fruit.

In 1887, seed of the wild Siberian crab apple Pyrus baccata was imported from the Royal Botanic Gardens, Petrograd, Russia, and sown at the Central Experimental Farm, Ottawa. Young trees grown from this seed were sent to the Experimental Farms at Brandon, Man., and Indian Head, Sask., in the prairie provinces, where the winters are very severe, the temperature at Indian Head falling at times to 50 below zero, Fahr. These trees proved quite hardy on a practically treeless prairie, while trees of cultivated varieties of crab apples and apples succumbed.

The fruit of this wild crab apple is very small, only half an inch in diameter, and it is quite astringent. In 1894 the late Dr. Wm. Saunders, then Director of the Experimental Farms, began crossing this wild crab apple with named varieties of apples in the hope of obtaining fruits of larger size and better quality than P. baccata but which would retain sufficient hardiness to endure the climate of the prairie provinces. All the crosses recorded have P. baccata as the mother; reciprocal crosses were not made. One hundred and sixty trees resulted from the first crossing and several hundred from subsequent work, or about 800 in all.

Some of the varieties of apples used as male parents are Tetofsky, Duchess, Wealthy, Anis, Beautiful Arcad, Broad Green, Excelsior, Fameuse, American Golden Russet, Haas, Herren, Krimskoe, McIntosh, McMahan, Osimoe, Pewaukee, Red Astrachan, Ribston, Scott Winter, Simbirsk No. 9, Swayzie, Tolman, Winter St. Lawrence and Yellow Transparent.

In 1899 thirty-six of the first crosses bore fruit and five of them were considered large enough and sufficiently good in quality to be propagated. By far the largest proportion produced fruit not sufficiently larger than the mother parent and of so inferior a quality as to be not worthy of propagation, but sixteen varieties were thought sufficiently promising to name. On weighing average specimens it was found that the best of these were from twelve to fourteen times heavier than the fruit of P. baccata. The largest fruits, however, were under 2 inches in diameter.

The better varieties of these crosses have little or no astringency and compare very favorably in quality with the named crab apples on the market. Nearly all of them retained the marked crab characteristics of long, slender stem; thin, tender skin, and crisp, breaking flesh.


After being propagated and thoroughly tested on the prairies some of these have proved hardier than any other varieties of apples or crab apples tested, thus marking a stage of development in hardy apples for the prairie provinces. Some of the hardiest varieties have proved to be jewel (P. baccata by Yellow Transparent, size 1.4 by 1.3 inches), Columbia (P. baccata by Broad Green, size 1.8 by 1.6 inches), Charles (P. baccata by Tetofsky, size 1.6 by 1.3 inches), Silvia (P. baccata by Yellow Transparent, size 1.4 by 1.5 inches), Tony (P. baccata by McMahan, size 1.6 by 1.4 inches), Elsa (P. baccata by Yellow Transparent, size 1.4 by 1.3 inches), Eve (P. baccata by Simbirsk No. 9, size 1.6 by 1.2 inches). Seedlings grown from these gave in nearly every case fruit smaller than the parent. As none of the fruits resulting from this cross was large enough to compare favorably with less hardy varieties of apples and crab apples, the best of these first crosses were, in 1904, recrossed with named varieties of apples with the object of obtaining varieties bearing larger fruits but which would retain sufficient hardiness to be grown in the open on the prairies.

In this work Dr. Saunders used the crosses as the mother parents in all cases. The varieties of apples used as male parents are McIntosh, Baldwin, Cranberry, Duchess, Northern Spy, October, Scott Winter, Simbirsk No. 9, Tetofsky, Yellow Transparent, Ontario, Gideon, Rideau, Haas, August, Walter, Wealthy, McMahan. From seeds obtained through this work 407 trees were grown at Ottawa which began to fruit in 1910 and of which a large proportion have borne. While many of these have borne fruit no larger than the mother parent, 24 have produced apples two inches and more in diameter. Some of the largest varieties which have fruited are Wapella (Dean by Ontario) size 2.25 by 2.25 inches; Angus (Dean by Ontario) size 2 by 2.5 inches. The parentage of Dean is P. baccata by Wealthy. Martin (Pioneer by Ontario) size 2.25 by 2.37 inches; Gretna (Pioneer by Northern Spy) 2 by 2.25 inches. The parentage of Pioneer is P. baccata by Tetofsky. Most of these second crosses retain the long, slender stem, the thin, tender skin, and the crisp, breaking flesh which are characteristic of Pyrus baccata, but a few are quite apple like.

It is not known yet whether these will be sufficiently hardy or not, but this will soon be determined.

It is to be regretted that the apple (Pyrus malus) was not used as the mother in these crosses, as it is believed by the writer that larger apples would have been obtained more quickly, but size might have been obtained at the expense of hardiness which is the first consideration on the prairies. If these second crosses prove hardier than any other apples or crab apples which have been tested they will mark another step in advance.


As some of the Russian varieties of apples had proved hardy in certain places in the prairie provinces and had produced considerable quantities of fruit, a new line of breeding hardy apples for the prairies was begun by the writer in 1912. Seed was sown of such hardy varieties as Anis, Anisette, Antonovka, Beautiful Arcad, Blushed Calville, Charlamoff, Hibernal, Tetofsky and Yellow Transparent. After the trees had made one season's growth in the seed bed they were transplanted one foot apart into nursery rows 3 feet apart on the six Experimental Farms at Brandon, Man., Indian Head, Sask., Rosthern, Sask.. Scott, Sask., Lacombe, Alta., and Lethbridge, Alta., and a few were sent to a sub-station at Fort Vermilion in the Peace River District. Some 50,000 trees were planted out in this way and it is planned to plant many more. Many of these trees have now passed through three winters and some of them have proved quite hardy, though a marked difference in this respect has been found. The hardy ones are now being transplanted to orchards for further test. It is hoped in this way also to obtain hardy varieties for Canada's coldest climates.

Actual size of the hardy Siberian crab apple (Pyrus baccata) which is being used by Canadian government breeders to cross with cultivated apples and produce a type that will be more resistant to the cold of the northern prairies. Some of the hybrids had fairly good flavor, without the astringency of the Siberian crab, but they lacked size, so they were recrossed with cultivated apples. The result is promising, in size and flavor, but it remains to he seen whether the hardiness of the Siberian stock has been retained. (Fig 7.)

There is a very large area in Canada where the apple succeeds well, but where the range of suitable varieties is limited as up to recent years Canada has depended mainly on foreign countries for her fruits, and many of the varieties introduced from warm countries have only been suitable for the most favored districts in Canada, hence an effort has been made to obtain other and hardier sorts which will cover the season better. As the Horticultural Division was not organized to do much work in cross-breeding, the writer, in 1898, believing that in an orchard at the Central Experimental Farm containing between 400 and 500 named varieties of apples all sorts of combinations of characters would be taking place by natural pollination and that the chance of obtaining some good varieties would be very great, had seed saved of some of the best flavored apples then fruiting in the orchard as well as some other varieties desirable on account of other characteristics. There were included in these McIntosh, St. Lawrence, Fameuse, Wealthy, Shiawassee, Swayzie, Northern Spy, Winter St. Lawrence, Langford Beauty, Scott Winter, Salome, Lawver, Gano and American Golden Russet.


The seedlings from these were planted in the orchard in 1901 and later until about 2,000 were set out, The results from this work have been very gratifying. The first tree to fruit from seed was a Wealthy seedling now called Crusoe which fruited in 1903, two years after planting and five years from seed. Detailed descriptions have been made of the fruit of more than 1,200 of these seedlings. There have been so many good apples among them that 100 varieties have been named because giving promise of being useful in some part of Canada. The male parent was unknown in this series of seedlings, but it is very interesting to note that a large proportion of such seedling varieties from McIntosh, Wealthy, and Northern Spy had characteristics strongly resembling the mother parent, while Fameuse, Swayzie, St. Lawrence and others were lacking in this respect, although in the case of Swayzie the spicy flavor of the mother parent was marked in most of the seedlings. Only about 5% of the seedlings have been small or crab-like. Further details in regard to these seedlings will be found in the reports of the Experimental Farms.

Following are the names of some of the best varieties:

McIntosh Seedlings.—Melba, Joyce, Pedro. These are three apples of the McIitosh type: the Melba, an August apple, the Joyce, a September apple, and the Pedro, an October apple, thus extending the season of apples of this type.

Northern Spy Seedlings—Autumn: Galton, Epsom, Thurso, Rocket, Tasty. Early winter: Lipton, Ascot. Winter: Elmer, Emilia, Sparta, Niobe.

While it is not claimed for any of these that they are better than Northern Spy or quite as good in most cases, they have all proved hardier than Northern Spy at Ottawa and they give a longer season of apples of the Northern Spy type.

The names might be given of seedlings of other varieties but as McIntosh and Northern Spy are two of the most popular varieties grown, their seedlings are given as examples. Detailed descriptions will be found in the annual reports of the Experimental Farms.

Previous to this series of seedlings, some 3,000 trees raised from seed received from north of Riga in Russia in 1890 had been tested but had given practically nothing of value as the fruit as a rule was of inferior quality.


A little work in cross-breeding was done in the Horticultural Division in 1895 when McMahan was crossed with Scott Winter and Waibridge with Northern Spy, but beginning in 1899 some work has been done almost every year since. The parents used in making crosses are Anis, Anisim, Antonovka, Baldwin, Baxter, Bethel, Bingo, Cobalt, Crusoe, Duchess of Oldenburgh, Dyer, Danville, Fameuse, Forest, Glenton, Gravenstein, Hibernal, Lawver, Lowland Raspberry, Malinda, Milwaukee, McIntosh, McMahan, Newton, Northern Spy, North Western Greening, R. I. Greening, Rosalie, Rouleau, Scott, Winter, Stone, Winter Rose, and Walton. Reciprocal crosses have been made in many cases. There have been two main objects in view in this work, first to obtain hardier winter apples for the colder parts of Canada where apples are grown commercially and, second, to obtain early bearing varieties covering the whole season, as there seems to be no good reason why more apples of the Northern Spy type should not be obtained which will bear as early as Wealthy and Wagener.

More than 1,000 trees are now growing as a result of a little crossing almost every year and nearly 100 of these have already fruited. So far not many apples have fruited which have been thought worthy of propagation, but there have been a few from a cross between McIntosh and Lawver where the object was to obtain varieties which would keep better than McIntosh.

In six out of ten crosses which have fruited with Lawver as the mother no marked resemblance to either parent is recorded, and similarly in three of the six with McIntosh as the mother. Of the four varieties with Lawver as the mother that have marked characteristics of the parent, two have distinct McIntosh flavor and two resemble McIntosh in color. The Lawver characteristics are not very marked. Of the six varieties with Mcintosh as the mother only two show marked resemblance to either parent in the important characteristics of color, flesh, and flavor, although as regards season a large proportion resembles both parents. The McIntosh seedlings from open pollination have given a larger proportion with marked McIntosh characteristics than has been the case in this cross. While there are none of the sixteen varieties of this cross which have yet fruited which are as good as McIntosh in quality, ten of the sixteen are better than Lawver in quality and thirteen of the sixteen are later in season than McIntosh, and most of the varieties are of high colour and attractive in appearance. Following are those which have been named: Lawver by McIntoshHolz, Vermac. Mcintosh by Lawver-Mavis, Rustler.

The new varieties of apples which have been referred to are being tested in different parts of Canada and no doubt some of them some day will take their place among the list of best varieties offered for sale, but their introduction is not being pushed as there are too many already.


Pears.—Some work has been done in recent years with pears. It has been found that certain Russian varieties such as Bessemianka and Gliva Kurskaya are comparatively immune from fire blight and these have been crossed with other and better varieties.

Plums.—Little cross-breeding has been done but many seedlings have been grown of Prunus americana and P. nigra, some of which have been named. It is believed that P. nigra offers the better field of work as it has more good characteristics for Canadian conditions than P. americana.

Cherries.—Seedlings are being grown of a wild cherry from North-Eastern Asia called Prunus tomentosa, the fruit of which varies considerably. This is a bush cherry which is hardy where the tree cherries do not succeed. Varieties with better fruit are sought.

Grapes.—Little progress has been made in breeding grapes though many seedlings of Rogers Hybrid grapes are now being grown and it is expected that some good sorts will be obtained.

Gooseberries.—Seedlings are being grown of crosses between Ribes oxyacanthoides, R. cynosbati, and R. grossularia varieties looking to obtaining larger fruited sorts not subject to mildew.

Currants.—A number of seedlings and cross-bred varieties, the best saved from a large number, are being tested out.

Strawberries.—A large number of seedlings have been grown but few good varieties have been obtained. Some of the most promising are Cassandra. Cordelia, Desdemona, Ophelia and Portia. In recent years crosses have been made between wild strawberries obtained from different parts of Canada and cultivated varieties with the object of obtaining hardier sorts.

Early varieties of vegetables are of great importance everywhere, but are much needed in certain parts of Canada. Selections for earliness have been made with tomatoes, beans, peas and corn particularly. The Alacrity tomato and Early Malcolm corn are two selections which have been most disseminated. During the past two years considerable work has been done in cross-breeding corn, the Squaw (flint), Early Adams (dent), and Early Malcolm (sweet) being mainly used as parents. The Squaw corn matures in districts where the nights are cool and the season without frost is short, whereas sweet varieties will not do so. It is hoped by crossing to obtain sweet varieties which will mature anywhere the Squaw does. Many interesting and promising crosses have been obtained.

Comparatively little work has been done in breeding ornamental plants, but some progress has been made with roses, sweet peas, geraniums, petunias and columbine. Two worthy varieties of roses originated at Ottawa by the late Dr. Wm. Saunders are Mary Arnott and Agnes, the former a brilliant crimson scarlet cross between Rosa rugosa and Prince Camille de Rohan, the latter, pale yellow with a salmon tinge, a cross between Rosa rugosa and Persian Yellow. Some interesting and attractive F2 seedlings are growing from a cross made between Berberis thunbergii and Berberis vulgaris purpurea.

There is a specialist in the Horticultural Division who devotes his whole time to plant breeding.