The rectangular objects are diatoms, with strands normally attached but capable of detaching from one another as seen here. Daniel Tecles' identification above is certainly a possibility; I am not knowledgeable about the specific biota of your region. All the background material in your photo seems to be just debris.
Daniel is right, it is Tabellaria flocculosa. The "zig-zag" bands are the result of divisions - one pole is still attached to the next cell via extracellular poysaccharides, the other not. This is the so.called side- or girdleband-view.
The organism of the first photograph seems to be a species of Navicula. In the second photograph it appears that there is an organism infected with Streptococcus sp., maybe with S.pyogenes.
For the diatoms identification, it would be better if you remove the organic material from the samples, by methods described in the book "Methods in stream ecology". I have done this and the identification got a lot easier, because then you can compare the photographies provided on the books from Krammer and Lange-Bertalot. The permanent slides are also a good way to identify to genus-species level and do not require a lot of reagents.