Introduction

Genetic improvement is central to the sustainable management of forest resources (^{Zobel and Talbert, 1984}), as it allows the obtainment seeds of high genetic quality and abundant quantity (^{Yuan et al., 2016}) from sexual or asexual seed orchards ―in the latter, with greater genetic gain, where controlled pollination can be carried out to cross individuals of high genetic value (^{Stewart et al., 2016}).

In order to establish asexual seed orchards, trees with phenotypic characteristics of economic interest are cloned; the most common method used is grafting (^{Muñoz et al., 2013}). Grafting techniques -mainly side veneer and terminal cleft- have been evaluated in the propagation of Mexican pines such as Pinus patula Schltdl. & Cham. (^{Aparicio-Rentería et al., 2013}; ^{González-Jiménez et al., 2022}), P. pseudostrobus var. oaxacana (Mirov) S. G. Harrison (^{Barrera-Ramírez et al., 2021}), P. engelmannii Carrière (^{Pérez-Luna et al., 2021}), and P. rzedowskii Madrigal & Caball. Del. (^{Solorio-Barragán et al., 2021}), with different results for each.

However, since in most of the grafting studies of Pinus species the survival rate has been less than 50 %, it is necessary to look for alternative techniques to increase this percentage (^{Pérez-Luna et al., 2020a}) in which the age and quality of the rootstock plant are considered as important factors (^{Castro-Garibay et al., 2022}; ^{González-Jiménez et al., 2022};). As for scions, the use of shoots from the last year of growth favors grafting (^{Pérez-Luna et al., 2021}). On the other hand, protecting the healing area with waxes or pastes strengthens the graft union and reduces the risk of tissue rotting (^{Muñoz et al., 2013}); however, this subject has been little studied (^{Pérez-Luna et al., 2020a}).

Pinus engelmannii, P. cooperi C. E. Blanco, and P. durangensis Martínez are species of the Ponderosae subsection distributed in northern Mexico and the southern United States of America, in the Sierra Madre Occidental (^{García and González, 2003}); these are also widely used for establishing commercial forestry plantations in northern Mexico, particularly in the state of Durango (^{Martínez and Prieto, 2011}).

Therefore, the objective of this study was to evaluate the effect of the grafting technique and the healing method on the survival of grafts of P. engelmannii, P. cooperi, and P. durangensis. The hypothesis was that there is a significant effect on graft survival in terms of species, grafting technique, and healing method.

Materials and Methods

Grafting site

Grafting was carried out in a 20 m long × 5 m wide (100 m²) and 2.4 m high greenhouse, with a white plastic cover 720-micron caliber, located at the Guadiana Valley Experimental Field of the Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP) in Durango, Mexico, at the coordinates 23°59’11.14’’ N and 104°37’20.54’’ W, at an altitude of 1 879 masl. A 60 % light retention shade net was placed over the plastic cover to reduce the incidence of solar radiation inside the greenhouse. Temperature (Figure 1a) and relative humidity (Figure 1b) were registered during the months of the experiment with an Elitech^® model Rc-4hc Data Logger thermohygrometer.

Figure 1 Monthly record of maximum, minimum, and average values of: a) temperature, and b) relative humidity. 

Grafting

Grafting was performed on January 27^th and 31^st, 2014, using scions with dormant buds, the phenological stage recommended for conifer grafting (^{Muñoz et al., 2013}). The terminal-cleft graft was made by cutting the rootstock stem at the height where its diameter coincided with that of the scion (Table 1), followed by a radial cut with a depth of 4 cm in the middle part of the cross section of the rootstock (Figure 2a). Two tangential cuts of 4 cm in length were made on the scion at the ends of the twig (Figure 2b), so that each cut of the scion ended with a slight inclination at the bottom to form a wedge that was placed and tied with the cut made in the rootstock plant (Figure 2c).

Figure 1 Grafting techniques: a) Transversal and longitudinal cut made to the rootstock plant; b) Cut in the scion for terminal-cleft grafting; c) Terminal cleft-graft joint with application of healing wax in its external section; d) Scion cutting for side veneer grafting; e) Side-veneer graft joint; f) Microclimate generated to maintain high relative humidity in the grafting area. 

For the side-veneer graft, a 4 cm long tangential cut was made where the diameter of the scion and the rootstock coincided, with a small 1 cm deep slit at the base of the cut (Figure 2d). A tangential cut of 5 cm in length was made in one of the scion veneers, and a second tangential cut was made in the veneer parallel to the base of the first cut, 1 cm long, to achieve a proper match with the slit-shaped cuts generated in the rootstock plant (Figure 2e). The grafts were performed by a single grafter.

The graft joint was tied with 4 mm glass rubber and sealed with white vinyl paint. In order to reduce the risk of pathogen contamination in the graft area, the paint was treated with Captan fungicide (Captan^® 50 WP) at a dose of 5 g L^-1. The high relative humidity in the grafting area was kept high with a transparent plastic bag containing water (microclimate) (Figure 2f). On top of the microclimate bag, a brown paper bag was placed to improve the temperature conditions and prevent direct solar radiation from affecting the graft. The rootstocks of the grafts were irrigated every other day with water and water-soluble triple 19 fertilizer (Poly feed GG^®) at a dose of 3 g L^-1. The brown paper bag was removed 30 days after grafting; then, at 45 days, an opening was made in the bags containing the water, and at 60 days the water was removed from the microclimate. Finally, after 70 days, the microclimate was completely eliminated.

Results and Discussion

Significant differences in graft survival between the study species were obtained at 60 days (p=0.003), 120 days (p=0.0002) and 180 days (p=2.7×10^-13). P. engelmannii grafts represented the first statistical group with 81.6, 55.4, and 32.9 % at 60, 120, and 180 days after grafting, respectively. At 180 days, P. cooperi and P. durangensis had 5.4 % and 0.8 % survival, respectively (Figure 3a).

Different letters indicate significant statistical differences between treatments according to Dunn's test. The bars represent the standard error.

Figure 3 Graft survival of Pinus engelmannii, P. durangensis and P. cooperi: a) by effect of grafting technique, b) by effect of species, and c) by effect of healing method. 

As for graft survival, significant statistical differences by technique effect were observed between the study species at 60 days (p=0.043), 120 days (p=0.04), and 180 days (p=0.036). For both techniques, there was a drastic increase in mortality between 60 and 120 days post-grafting, as survival with both the terminal-cleft and side-veneer techniques decreased from 63.3 to 29.7 % and from 73.6 to 42.5 %, respectively. At 180 days, the highest survival rate was obtained with the side-veneer technique with 21.3 %, while with the terminal-cleft technique, the rate was 9.7 % (Figure 3b).

Significant statistical differences were also registered in the effect of the healing method at 60 days (p=0.011), 120 days (p=0.03) and 180 days (p=0.013); although there was high mortality between evaluations in both healing methods, the application of healing wax favored graft survival (76.3, 41.3 and 16.9 %, at 60, 120 and 180 days, respectively), compared to natural healing (without wax), with which the rates obtained at the same times were 60.8, 30.8 and 9.1 % (Figure 3c).

Several studies indicate that the best grafting technique for propagating species of the Pinus genus is the end-grafting technique. With this technique, ^{Almqvist (2013)} observed 84.7 % survival in P. sylvestris L., and ^{Pérez-Luna et al. (2021)}, 80 % in P. engelmannii. Likewise, ^{Castro-Garibay et al. (2022)} cited 89 % survival rate in P. greggii var. australis Donahue, Jeffrey K. & López A., R. While, with the lateral plating technique, survival in P. rzedowskii was above 80 % (^{Solorio-Barragán et al., 2021}), and 73 % in P. patula (^{González-Jiménez et al., 2022}). On the other hand, ^{Aparicio-Rentería et al. (2013)} and ^{Barrera-Ramírez et al. (2021)} estimated 29 and 30 % survival rates in P. patula and P. pseudostrobus var. oaxacana, respectively.

In this study, survival rates with both techniques were low: below 10 % at 180 days in terminal cleft grafts, and 21.3 % with side veneer. The above may be attributed to factors unrelated to the object of evaluation in the research; one of these factors is the age of the rootstock plant, given that authors such as ^{Castro-Garibay et al. (2022)} and ^{González-Jiménez et al. (2022)} recommend the use of rootstock plants no older than one year of age. However, in certain slow-growing species, it is complicated to use rootstocks of that age, because by that time these do not reach the same growth in diameter as the bud to be grafted. Particularly P. engelmannii, which exhibits a cespitose state (^{Pérez-Luna et al., 2019}), and for which three-year-old rootstocks are required. On the other hand, ^{Pérez-Luna et al. (2020b)} suggest keeping the grafts at temperatures no higher than 30 °C, which was exceeded in the present study (Figure 1a).

The large difference in survival in P. engelmannii, compared to P. cooperi and P. durangensis is indicative of the greater suitability of P. engelmannii for propagation by grafting. In this regard, ^{Hibbert-Frey et al. (2011)} note that not all conifer species can be successfully cloned by grafting. Also, it is documented that there are species that are susceptible to better propagation by rooting cuttings, such as P. leiophylla Schiede ex Schltdl. & Cham. (^{Cuevas, 2014}) and Abies fraseri (Pursh) Poir. (^{Rosier et al., 2005}).

On the other hand, the effect of healing agents has been little studied in conifer species. ^{White et al. (1983)} and ^{McKeand et al. (1987)} obtained 83 and 85 % survival in Pinus taeda L. by applying paraffin and resin-based wax at the graft junction, respectively. ^{Blada and Panea (2011)} in Picea pungens var. glauca Regel registered 41 and 87 % survival rates in paraffin and eco-wax grafts, respectively.

Regarding the effect of the interactions, significant differences (p<0.05) were observed in the interaction between species and the grafting technique factor in the three evaluations, with different levels of success, achieving more favorable results in P. engelmannii with both grafting techniques and a final survival rate of 40 and 25 % in side-veneer and terminal-cleft grafts, respectively. The second group included the four interactions with the two techniques (terminal cleft and side veneer) for the species P. cooperi and P. durangensis (Table 2).

In the interaction between the species and the method of healing, significant differences (p<0.05) were obtained in the three evaluations; the interactions were divided into four statistical groups. The largest group corresponded to P. engelmannii grafts with healing wax, with a final survival rate of 43.3 %. Statistical differences (p<0.05) were only detected at 60 and 120 days in the interaction between the grafting technique and the method of healing, with the best results at 60 days, when healing wax was applied, and at 120 days with the side-veneer technique, regardless of the type of protection (Table 2).

^{Chávez (2016)} evaluated terminal-cleft grafting of P. engelmannii, P. cooperi and P. durangensis, and cited 20, 34, and 0 % survival, respectively, three months after grafting; the result obtained in this work confirms how difficult it is to propagate the species evaluated, at least with the methodology employed.

The triple interaction exhibited significant statistical differences (p<0.05) in all evaluations. The most favorable interaction in the three evaluations occurred in P. engelmannii at 60 days with the use of healing wax, with both grafting techniques, at 120 days only the dominance of the species prevailed, and at 180 days, the side-veneer graft with application of healing wax stood out. The results with lower survival corresponded to P. cooperi and P. durangensis, regardless of the factor, which indicates that not all species respond in the same way to grafting (Table 3).

The effect of healing waxes on graft survival varies according to the species studied (^{Ungureanu et al., 2013}). For example, ^{Blada and Panea (2011}; ²⁰¹²⁾ studied the effect of the application of wax based on bioregulators in the grafting zone in the grafted area in Picea pungens var. glauca and Abies concolor (Gordon & Glend.) Lindl. ex Hildebr.; their results indicated a survival of 87 and 96 %, respectively, at five months of age of the grafts, which were maintained under greenhouse conditions at an average temperature of 25 °C.

In the three assessed species, the highest survival was obtained in side-veneer grafts, which may be indicative of a greater ability of the grafter to develop this technique. However, it is possible that the conifer species studied have more affinity to be propagated by other propagation methods.

Conclusions

The species with the best response to grafting is P. engelmannii; side-veneer technique and the use of healing wax increased graft survival. In the case of P. cooperi and P. durangensis, it is recommended to evaluate other factors that could favor the survival of grafts, or otherwise, explore other methods of vegetative propagation.