Brodiaea santarosae
The Santa Rosa Basalt Brodiaea

Tom Chester, Wayne Armstrong and Kay Madore

Distinguishing Characteristics of Brodiaea santarosae

Brodiaea santarosae is easily distinguished from other Brodiaea species in southern California by its large flowers and distinctive, variable staminodes; see the pictorial key.

We did not include B. elegans or B. kinkiensis in the analysis here since they do not come into the geographic range of any southern California Brodiaea except for B. terrestris ssp. kernensis, and we therefore didn't sample them for this analysis.

Detailed morphological analysis has found eleven separate characteristics that distinguish B. santarosae from the two species with which it has previously been confused, B. filifolia and B. orcuttii. Five of those eleven characteristics also distinguish B. santarosae from B. terrestris ssp. kernensis.

There is a distinct pattern to most of these differences:

Brief summaries of the differences are given below; the bold name for each characteristic is linked to a page with more information, including plots.

The distinguishing characteristics are:

  1. Habitat. B. filifolia and B. orcuttii are confined to the wettest areas. B. terrestris ssp. kernensis grows in many habitats, but generally does not occur in the wettest areas. B. santarosae grows in many habitats as well, including even wetter areas next to vernal pools, but can also grow in drier locations than even B. terrestris ssp. kernensis.

    We also found a difference in scape (peduncle) lengths, but suspect this is related to the height of the surrounding plants in each habitat, and so do not list it as an independent difference. Both B. filifolia and B. orcuttii generally have short scapes, whereas B. santarosae and B. terrestris ssp. kernensis generally have long scapes.

  2. Inflorescence bract lengths. There is a dramatic difference in the length of the bracts surrounding the base of the pedicels. 65% of the bract lengths for B. filifolia and B. orcuttii were less than 6 mm, but none of the bract lengths for the other two species were that short.

  3. Pedicel lengths. B. filifolia and B. orcuttii have short pedicel lengths. B. terrestris ssp. kernensis has a histogram of pedicel lengths shifted only slightly to larger values, whereas B. santarosae has a histogram shifted dramatically to larger values.

  4. Perianth tube length. The perianth tube of B. santarosae is 30% longer, and the tube of B. terrestris ssp. kernensis is 60% longer, than the average tube of B. filifolia and B. orcuttii.

  5. Perianth lobe length. The perianth lobes of B. santarosae and B. terrestris ssp. kernensis are 34% longer than the average lobes of B. filifolia and B. orcuttii.

  6. Ovary length. The ovaries of B. santarosae and B. terrestris ssp. kernensis are 50% longer than the ovaries of B. filifolia and B. orcuttii.

  7. Anther lengths. The anthers of B. santarosae are 30% longer than the average anthers of the other three species.

  8. Style lengths. B. santarosae has styles that vastly exceed the lengths for the other three species, being 64% longer. A single observation of the style is usually enough to reliably identify a flower of B. santarosae.

  9. Staminode properties. Differences in staminodes are the hallmark of Brodiaea species, and are used heavily in the key to distinguish them. B. orcuttii is distinctive in having no staminodes in any of its flowers. B. terrestris ssp. kernensis is distinctive by its long, purple rectangular staminodes. B. filifolia is distinctive in having intermediate-length staminodes tapered to the tip and reflexed against its perianth. B. santarosae is distinctive in having staminodes recurved to erect, and so variable in length that individual flowers often show large differences in their three staminodes.

  10. Lack of scaling of perianth tube with perianth length. This is discussed immediately below.

  11. Lack of scaling of ovary with perianth length.

    Most species have flowers in which all the floral parts grow larger as the perianth grows larger; our analysis of seven flora parameters showed only two exceptions to that rule. Surprisingly, the perianth tube and ovary length for B. santarosae did not scale with perianth length. It is tempting to say that these values have simply reached their maximum values for the large flowers of this species, but both characteristics grew with perianth length B. terrestris ssp. kernensis, which had equally large flowers.

    Although the fitted slope of the staminode length for B. santarosae was actually negative with perianth length, indicating a tendency for the maximum staminode length to decrease with increasing perianth length, we could not confidently make any conclusion here due to the large variation in staminode lengths.

    The second exception were that the staminodes of B. terrestris ssp. kernensis did not scale with perianth length.

    We have also done a Principal Components Analysis (PCA) on seven of the floral parameters discussed above. The PCA describes the total variation and correlation in these measurements, and helps to see how well the measurements cluster into species. The PCA shows four well-separated species, with B. santarosae well separated from the other three species. It also strongly supports our findings of true F1 specimens of B. filifolia X B. orcuttii hybrids at San Marcos.

Go to:

Copyright © 2007 by Tom Chester, Wayne Armstrong and Kay Madore.
Comments and feedback: Tom Chester
Last update: 16 October 2007