Capillary zone electrophoresis (CZE) of DNA 23.1 to 48.5 kb in length in polyacrylamide solutions of several concentrations provides evidence for polymer concentration and DNA length-dependent stretching and orientation of these species and suggests an effective separation at a polymer concentration of about 0.6%. Applying a 0.1% polyacrylamide concentration to the lambda-phage DNA ladder, at least 5 components are separated; separation improves with lowering of the field strength to 2 V/cm and, correspondingly, extended duration of CZE. Saccharomyces pombe chromosomal DNA separates into 3 major components on CZE at high field strength (270 V/cm) in 0.9% polyacrylamide solution, confirming a previous finding made on electrophoresis in a 1.1 mm ID tube at low field strength. However, the finding is limited to one source of the DNA plug, and the chromosomal identity of the components remains unknown. Methodological problems in the CZE of large DNA relate to the need for extended duration of pressure injection if absorbance detection is applied, the need to define the starting zone after extended pressure injection, the need to melt and digest agarose plugs prior to loading, and related needs for thermostating of the sample chamber and for software compatible with low voltage operation.

DNA electrophoresis in gels and solutions of agarose and polyacrylamide was objectively evaluated with regard to separation efficiency at optimal polymer concentrations. In application to DNA fragments, polyacrylamide gels were superior for separating fragments of less than 7800 bp, and agarose gels are the best choice for larger fragments. Agarose solutions are nearly as good as polyacrylamide gels for small DNA (< 300 bp). Agarose solutions have a higher efficiency than polyacrylamide solutions for DNA of less than 1200 bp. Separation efficiency sharply decreases with increasing length of DNA. Retardation in polyacrylamide solutions was found to depend on polymer length in a biphasic fashion. The choice of resolving polymer concentrations depends on the progressive stretching of DNA in proportion to polymer concentration. The rate of that stretching appears higher in polyacrylmide solution than in gels or in liquid or gelled agarose. Application of polymer solutions to capillary electrophoresis raises further problems concerning agarose plugs, DNA interactions with the polymers, operation at low field strength and long durations as well as detection sensitivity.

Agarose gel electrophoresis has been shown to give rise to non‐linear plots of log (mobility) vs. gel concentration of spherical viruses (Serwer,[1]) and cellular vesicles (Gottlieb et al.[2]). This finding also applies to proteins as shown in this study. Considering that in the non‐linear plot, the slope becomes a function of gel concentration, it is possible to determine particle properties and gel parameters by a modification of the conventional method derived from the Ogston theory for long‐fiber gels. This treatment shows: (a) In application to data obtained from gel electrophoresis (0.4 to 1.6 % agarose) of viruses (13 to 42 nm radius) and with increasing gel concentration: (i) an increase of apparent total fiber length per g agarose matrix, (ii) a reduction of apparent fiber radius and (iii) a constant fiber volume (per g matrix material) of the agarose fiber. The values of the fiber radii identify the fibers as agarose supercoils (or aggregates of it) with a radius of 20–55 nm. (b) In application to data obtained from gel electrophoresis (1.2 to 8 % agarose) of proteins (1.7 to 5.8 nm radius): a fiber volume indicative of additional sieving by the agarose double helix (of known radius of 0.5–2 nm). This is in agreement with a previous suggestion by Serwer [1] that proteins are able to penetrate into the double‐helical network of which the agarose supercoil consists. (c) The likelihood of a continuous transition from case (a) to case (b).