If freezing limits establishment of warm desert shrubs at high latitudes, shrubland distributions may be altered as a result of rising global temperatures. However, variation in plant physiology and morphology can be observed across climate gradients and may be acted on by selection to produce adaptation to local climate conditions, thereby ameliorating low temperature stress. Freezing damage in evergreens is closely linked to vessel size distribution because larger xylem conduits are more likely to become air-filled during freezing. In addition, plastic variation, rather than genetic, may be responsible for differences in freezing tolerance among populations. In order to determine if local adaptation to freezing is present in two species of the genus Larrea, L. tridentata and L. divaricata, we investigated xylem vessel size distributions in field grown L. tridentata adults and saplings grown in a common garden from high latitude (Sevilleta National Wildlife Refuge) and low latitude (Higuerillas, Mexico) sites in the Chihuahuan Desert in North America. High latitude (Bajada del Diablo, Argentina) and low latitude (Chamical, Argentina) populations of Larrea divaricata were selected for investigation from the Monte Desert in South America.
Field sample collection: Seed and wood samples were collected at two North American sites in the winter of 2006 and at two South American sites in the winter of 2008. Wood samples were wrapped in plastic to reduce dehydration. Following their return to the lab, wood samples were placed in a freezer until they could be analyzed. Seeds were sown within three weeks of collection.
Plant propagation: Seeds were germinated in the University of New Mexico research greenhouse within one month of collection and received only natural light. Seeds were sown in flats containing a 4:1:1 mixture of sand, peat, and perlite, then individual seedlings were transplanted to 1 L containers within seven days of germination. Seedlings were kept well-watered and received spray irrigation once or twice daily depending on temperatures and evaporative demand. One teaspoon of slow release fertilizer (Osmocote 14:14:14) was applied immediately after and every four months following transplanting.
Xylem vessel size distributions: Both field and lab samples were sectioned and imaged in June 2009. A total of 44,672 vessels from 72 branches were counted (eight branches from each field and greenhouse population). Wood was cut into sections 26 micrometer thick using a sliding microtome. Sections were stained with a 70% solution of toluidine blue and wet-mounted on a slide. Sections were magnified 100x and imaged using a digital camera attached to a dissecting microscope.
We defined four rings of equal diameter and took a single image of each ring. We counted all the vessels contained in the images of the four rings moving from the edge to the center of the section, making sure not to include individual vessels in more than one image. For each day that images were generated we also made an image of a micrometer scale. The image of the scale was opened in ImageJ and the function "Set Scale" was used to determine the number of micrometers/pixel. Images of xylem were then analyzed in ImageJ using the appropriate scale. They were inverted, then thresholded so that only the vessels were represented by colored pixels. To verify that the particles measured by ImageJ represented actual vessels a drawing of the measured particles, generated by the software, was compared to the original .jpg image.
When objects appeared in the thresholded image that did not correspond to intact vessels appearing in the original image, such as small cracks in the section or occasional torn vessels, these objects were manually erased and the analysis was redone. The image was analyzed using the "Analyze Particles" function, which determined an area and the x and y diameters of a best fit ellipse for each individual xylem vessel. We excluded particle measurements with shapes touching the edges of the image, exhibiting a circularity less than 0.1, or exhibiting a major diameter of the best fit elipse less than 7.5 micrometers, which corresponds to the lower size limit of vessels measured by Martinez-Villalta and Pockman (2002).
Instrument Name: Sliding Microtome.
Manufacturer: American Optics.
Model Number: 860.