NID

Field Methods:
Beginning in May 1998, monthly samples were collected from 3 sites along the Rio Mameyes. After Hurricane Georges on September 21, 1998, 2 more sites were added and samples were collected weekly for the following 10 weeks. In January 1999 monthly sampling was resumed.

On each date 9 samples were collected at each site in the wide, deeper sections of the river, and 3 samples were collected from sites that were shallow and narrow. In the wide sections of river, 3 samples each were collected from the right, center, and left of the river. Each sample is represented by the scrapings of a known area of one rock (16 or 32cm2) using a template, a razorblade, a toothbrush, and deionized water to rinse the sample into a labeled bottle. Samples were collected from the same depth on each date (45 cm deep on the left and right sides of the river and 80 cm deep from the center of the river). In the shallower sections of the river samples were collected from approximately 10 cm depth. For more details see Report)

Lab methods:

Samples were kept cool and transferred to the lab where they were preserved with 1-2 ml of Lugol's solution and 1-2 ml of 10% formalin. After at least 8 hours to allow for settling the samples are reduced, volumes recorded, split, and transferred to 25 ml scintillation vials. Part of each sample was saved for later ash-free dry weight analysis and part was saved for enumeration and identification.

Ash-free dry weight:

Preparation for ash-free dry weight was performed according to Standard Methods for Wastewater Analysis. Glass-fiber filters (0.45um) were dried, burned at 450oC, and weighed. Each sample was filtered with suction using a pump. Samples were rinsed thoroughly with DI water to remove traces of formalin and Lugol's solution. (A test was conducted to see if residual Lugols or formalin was left on the samples after incineration at 450oC with and without rinsing. A significant amount of Lugols remained on samples without rinsing, but was not detected after rinsing.) The sample-laden filters were dried thoroughly for 24 hours at 105oC, weighed, burned for 6 hours at 450oC in a muffle furnace, then weighed again. The difference in weight represented the organic component of the sample, i.e. algae, bacteria, detritus. This value was converted to grams per meter2 of river substrate.

Enumeration:

A 1 ml aliquot of the fraction reserved for this analysis was placed in a Sedgewick-rafter counting cell. This was examined at 20X for live and dead cells. The proportion of live to dead cells seen in one vertical strip of the Sedgewick-rafter cell was recorded. In more densely crowded samples a convenient number of fields were scanned (usually 9 to 15 fields). The volume of the sample, the volume of the aliquot analyzed (the fields or the strips), the area of the substrate scraped (a fraction of the whole sample), and the number of cells counted were used to calculate the number of cells per mm2 of river substrate. Estimates of the area covered by each cell were also made.

Identification:

The enumerated aliquot was returned to the sample in a glass scintillation vial. The contents of the vial was then settled for at least 8 hours, reduced via siphoning, resuspended with deionized water, and resettled at least three times to remove preservatives. After the last reduction, an equal volume of nitric acid was added to the sample along with a few grains of potassium dichromate (Patrick, 1967). This was settled for at least 8 hours to dissolve the organic component of the sample and leave only the siliceous shells of diatoms. This was then diluted with DI water, settled, siphoned, and diluted again at least 5 times to return the sample to a neutral acidity. A few drops of shaken sample (not too concentrated to obscure microscopial viewing) was transferred to a #1 cover slip that had been labeled with the sample ID, and dried on a hotplate. A drop of Cargill Meltmount was melted onto a labeled microscope slide and the sample-crusted cover slip was inverted onto this drop and gently pressed flat, when necessary. Slides prepared in this way could be examined at 100x for the fine detail necessary for taxonomic identification. Approximately 900 cells were counted from each site (100-300 per slide). The number of species per number of cells counted was used to determine the diversity present at each site.