Grafting Magnolia on Liriodendron

"the union is not possible, " though I By R.B. Figlar don't recall anything more specific from our discussion. Literature Ever since becoming interested in searches by Peter Del Tredici (Arnold magnolias about ten years ago, I've Arboretum), by Charles Tubesing been fascinated with the taxonomic (University of British Columbia), by link between Magnolia and Lola Koerting (Brooklyn Botanic Liriodendron, both of which, of course, Garden), and by me failed to turn up are members of Magnoliaceae. Here in anything. So with no reports of the northern New Jersey-southeastern successes or failures of previous New York area, Liriodendron tulipifera attempts, I began the experiment. is one of the more common indigenous Magnolia acuminata was chosen as trees, occurring in nearly all forest the flrst guinea pig for several reasons. habitats. The growth habit as well as the wood When our house was built in 1980. of the cucumbertree is considered the disturbed areas along the perimeter similar to that of the tuliptree, at least of our property became prime breeding by foresters (Collingwood & Brush, grounds for L. tulipifera seedlings. 1974). Moreover, on the basis of floral Though tuliptree saplings can often be odors (Thien, Heimermann, & a nuisance in the garden, rather than Holman, 1975) there is a high degree of removing them we decided to try correlation between the odor of L grafting magnolias onto these young tulipifera and M. acuminata. Thus, it volunteer tuliptrees. If the close was presumed that perhaps M. relationship between Magnolia and acuminata was more closely related to Liriodendron allows grafting between L tulipifera than were other species of these two genera, our little magnolia. Liriodendron "weeds" could become Using ordinary chipbud grafting quite useful. methods (McDaniel. 1973), I grafted Cases of successful graft "takes" M. acuminata buds on various L between different genera within a tulipifera saplings in our backyard in family are not commonplace. Trifoliate 1980, 1981, and 1982. Of the total of orange (Poncirus trifoliata) has 10 chipbuds, all failed to take. During sometimes been used as an understock the same period "control" chipbuds of for commercially grown orange trees, while quince (Cydonia oblonga) has been used as an understock for pear and loquat (Hartmann & Kester, 1983). Even within Magnoliaceae, successful unions have been reported between Magnolia and Michelia (Savage, 1973) and between Magnolia and Manglietia (Chiappino 1975, shown on cover of AMS Newsletter, Vol. XI, No. 1.) What about grafting between Magnolia and Liriodendron? Joe McDaniel back in the late 1970s M. tripetala grafi on Liriodendron, indicated to me in a conversation that bud now two inches long. 21 GROWTH OF M. TRIPETALA GRAFTED ON L. TULIPIFERA va GROWTH OF M. TRIPETALA ON ITS OWN ROOTS (Control Plant)

Co tr 1 91d sho t Co t ot 7 I o1

0609th (Tnchoot

2 6. D

91' Tt t d orr d lro 1 7n 2.

2

7. 0 — 9. 0

6/6 6/13 6/20 6/27 7/9 7/TT 7/Ts DATE &ttonth/Ooy&

L tulipfera on L. tulipifera and of M. leader of a 7-foot L tulipifera sapling, acuminata on itself were successful while the two dormant buds were more than 50 percent of the time. grafted on the lower auxiliary branches Thus, even though 10 was a small —one on the same tuliptree and the sample size, the three different years of other on a second sapling. The buds trials and the results of the control seemed to heal nicely and looked good grafts made a pretty good case for through the rest of the fall and winter. wasn' ruling out the potential for a successful Despite these good signs, I t union between the cucumbertree and really very optimistic yet. the tuliptree. Spring of (984 came late and trees in I had just about given up, then, our area didn't begin to break bud while inspecting the ripening wood of a until late ApriL Some magnolias, 2-year-old M. fraseri in our backyard, I including our permanent plant of M. was intrigued by the striking tripetala (the scion donor), began to resemblance of the word of M. fraseri leaf out by the fourth week of that to the stems of L. Tulipifera. Why not try month. The experimental grafts had this Magnolia on Lirodendron? Instead not. Though the buds still looked of butchering my small M. fraseri for healthy, I still wasn't really expecting budwood, I decided to take scions of anything to happen. M. tripetala, since it and M. fraseri After coming home from work, while are considered closely allied in section taking my usual walk through the yard Ryti dosperm um. (April 29), I was excited to see that one In early September l983 I grafted of the experimental buds was beginning three M. tripetala buds: one well to show new growth! By the next day developed (hook shaped) ~//2-inch bud, all three buds showed evidence of and two of the usual small dormant growing out. Within a few weeks the buds. The well developed bud was topworked bud really began to take topworked five feet up the central off. By June 6 it had grown 3 inches 22 and showed 6 leaves (nodes). By wood did remain on the Liriodendron comparison, the control plant (scion stock but there were no buds on it. donor) of M. tripetala. which had an Analysis. Despite the excessive callus earlier start, had also grown three accumulation. the small necrotic inches but with 9 leaves on its terminal fissure. and the nearly clean break at shoot. In contrast to the performance the union, the growth performance of of the topworked bud, the two the graft of M. triperala on L auxiliary set buds were still in the tulipifera was astonishing: process of expanding and were no ~ The graft grew a total of 28 inches. more than 3/4-inch long. I soon lost This was two times more than any interest in these two buds and magnolia in our yard and almost three neglected them thereafter. times more than the growth of the During June the graft began to grow control plant of M. tripetala. quite vigorously. Except for my ~ During the five weeks ending July I I occasional trimming back of the graft grew at the rate of five inches Liriodendron growth from other per week. This compares to I.7 inches branches on the stock, the M. tripetala per week for the control plant. graft was left to proceed on its own- ~ The graft put out an average of l. 7 no fertilizer and no staking or splinting new nodes per week during the same of the graft to the stock for additional five-week period, compared to only I.O support. The union healed nicely with per week for the control plant. abundant callusing (see closeup photo). ~ The leaves on the graft were larger During the latter part of June one (up to 19 inches long) and of a darker side of the callus mass developed a green than those on the control plant. small brown line (necrotic layer). Some ~ The graft shoot developed seven sap appeared to be leaking from the auxiliary branches ranging from I inch area a few days later. Fortunately the to 7.5 inches long. The control plant problem didn't get any worse and in developed only two auxiliary branches. fact the necrotic line seemed to heal ~ Though it grew for only ten weeks. over after a week or two. the weight of the graft after it blew By early July it became evident that down was I pound. the callus tissue was continuing to Conclusion. This experiment with accumulate at the point of union grafting Magnolia on Liriodendron between the M. tripetala scion and the does prove that an initial union tuliptree stock. Usually by this time the between the two genera is at least callus bridge should begin to possible. The evidence (as shown by differentiate into new xylem and phloem tissues. These would then provide a normal vascular connection between scion and stock (Hanman gt Kester, l983). Apparently, though, this had no adverse effect whatever on the growth rate of the graft, which was now putting on live inches of new terminal growth a week. (see graph). Then fate stepped in, and during the late evening of July I I a cold front moved through the area. accompanied by winds of 20 to 25 miles per hour. The next morning the graft was found Closeup view of M. tripetala (vt) lying on the ground, having been grafted onto Liriodendron tulipifera blown down by the strong winds. The (B). /vote extensive callus buildup at break was more or less clean at the graft union. Diameter of stork at B: point of union. Some M. tripetala l/2 inch. 23