The amount to load per well is variable. What is most important is how much DNA there is in the bands you wish to resolve. The least amount of DNA that can be consistently detected with ethidium bromide is about 10 ng. The most DNA you can have in a band and still get a sharp, clean band on an ethidium bromide stained gel is about 100 ng. These amounts will be less on gels stained with more sensitive stains such as GelStar™ Stain. On a GelStar™ Stained Gel it is possible to detect as little as 20 pg dsDNA. The optimal amount of DNA to load in the well is calculated by the fraction of the total DNA which is in the band of interest. If you are unsure of how much DNA is present, load varying amounts in several lanes if possible. The optimal amount of DNA to load in the well may be calculated by the fraction of the total DNA which is in the band of interest, represented by the following: Fragment of interest (kbp) divided by total size of DNA sample (kbp) X 100 = % DNA in band of interest NOTE: The most DNA compatible with a clean sharp band is approximately 100 ng. For example: The size of your DNA sample is 48.5 kbp and when run on the gel 8 fragments are separated. Your fragment of interest is 2.3 kbp. Calculation: (2.3 kbp/48.5 kbp) X 100 = 4.7% DNA in fragment of interest If you load 1 ?g of DNA, then 4.7% of the 1 ?g of loaded sample will appear in your fragment of interest (47 ng). To further increase the sharpness of the bands, use a Ficoll™ Based Loading Buffer, such as Lonza DNA Loading Buffer (Cat. No. 50655) instead of sucrose-based or glycerol-based loading buffers. The use of lower molecular weight glycerol will allow DNA to stream up the sides of the well before electrophoresis which results in U-shaped bands. Loading buffer that is too high in ionic strength can cause the bands to be fuzzy. In the ideal situation, the DNA sample should be suspended in the same solution as the running buffer. If this is not possible, use a sample buffer with a lower ionic strength than the running buffer.