HORTICULTUKK. 259 



early winU'r wliik' tlu- cutting plants prodnc-ed the ln'avii'Ht hlodiu later in tlu' season. 

 The question as to which method is best will therefore depeml upon the time the 

 flowers are in greatest demand. 



As a result of these experiments it is believed that, where bloom rather than stock 

 plants is desired, the test is emphatic in showing the desirability of plants propagated 

 from flowering wood. The cumulative effects of propagating roses from blind wood 

 or from flowering wood year after year is not marked. 



The proijag-ation of the Easter lily from seed, (i. W. Olivek ( U. S. Dcjd. Ayr., 

 Buretiu of I'Uiiit Indudry Bui. 39, pp. 24, ph. 7). — The author advocates the growing 

 of Easter lilies from seed rather than from bulbs as one method of avoiding the 

 troublesome lily diseases. Reproduction from seed is sho^yn to be entirely practica- 

 ble and as rajiid as from ])ulbs. Illustrations are given of seedling lilies in bloom 6 

 mouths and 13 days after germinating. Detailed directions are given for i)ollinating 

 flowers, sowing the seeds, and the future care of plants. 



Fertilizer experiments in 1899 with Chrysanthemum indicum, variety 

 Viviand Morel, M. Hoffmann {Gartenflum, 5:2 [1903), Xos. 11, pp. 297-305; U, pp. 

 315-319; 13, ji/). 347-352, figs. 6). — Pot fertilizer experiments with chrj^santhemums 

 were made at 4 different stations in Germany and the results obtained are here 

 detailed. A soil of fair quality was used in each experiment. One pot was left 

 unfertilized as a control. Fertilizers were added to 4 other pots in the liquid used 

 for watering as follows: (1) Sulphate of ammonia in the proportion of 1 to 200; (2) sul- 

 phate of ammonia in the proportion of 1 to 50; (3) licjuid manure in the proportion of 

 1 to 10; and (4) liquid manure in the proportion of 1 to 10, bone meal 1 to 200, and 

 sulphate of ammonia 1 to 2,000. The experiments covered a period of 234 days. The 

 plants were transplanted 4 times. Careful data were taken of the height of the jilant, 

 number of branches, color, -width and length of the leaves, total weight of all the dif- 

 ferent parts at maturity, and the nitrogen content of the best of the largest and of the 

 smallest roots, branches, leaves, and flowers. These data are all tal)ulated. 



In general the nitrogen content of the leaves and flowers was about 3 times as 

 great as the nitrogen content of the roots and branches. The best plants in these 

 experiments were ol)tained by the use of suliihate of ammonia in the proix)rtion of 

 1 to 50; following this came the plants fertilized with sulphate of ammonia in the pro- 

 portion of 1 to 200, and then those fertilized with liquid manure to which bone meal 

 and sulphate of ammonia had been added. 



The book of the daffodil, S. E. Bourne {London and Xnr York: John Lane, 1903, 

 pp. XIJ ^r 112, ph. IS, figs. 4)- — Cultural d'rections are given for the daffodil or 

 narcissus, with descriptions of a large numl)erof different varieties and notes on how 

 to form a collection, exhibit specimens, and judge them at shows. This is the six- 

 teenth of the series of Handbooks of Practical Gardening edited by H. Roberts. 



Spraying to kill pond scum, H. P. Hedeick {Gardening, 11 {1903), No. 259, p. 

 295; Amer. Florist, 20 {1903), No. 785, p. 750). — The green matted growth of low 

 water plants which is frequently found in fish ponds, lily ponds, and water gardens, 

 and which is usually got rid of by raking off and carting away, was destroyed by the 

 author by spraying with Bordeaux mixture. The mixture was first applied full 

 strength about August 1; 24 hours afterwards the green scum had turned brown and 

 in another day had sunk to the Ixjttom of the jiond and the water was perfectly 

 clear. Water lilies were slightly injured by the spray, showing that the Bordeaux 

 mixture was a little too strong. 



When the Bordeaux mixture was applied al)out J full strength the same result 

 was obtained with the green scum, with no bad effects to the lilies. After the first 

 spraying it was about a month before a second was required. Fish and frogs in the 

 I)ond did not appear to be injured by the spraying, while it is believed that tb.e 

 breeding of mosquitoes was materially checked. The author recommends a spray 

 composed of 4 lbs. of copper sulphate, 4 lbs. of unslaketl lime, and 00 gal. of water. 



