Invertebrate drift was sampled monthly from May through September of 1979 in thirteen subarctic, Alaskan streams. Samples were netted continuously for twenty-four hours to eliminate the time-of-day variation. Station mean values of drift concentration and export were regressed stepwise with chemical and physical parameters expected to be predictive. Alkalinity and average velocity together explained over 90% of the variation in drift export rates, expressed either as weight or numbers per day. These two factors plus discharge explained over 90% of the variation of drift concentrations as numbers per cubic meter;The importance of stream alkalinity in the prediction of drift has been shown previously in a study of seven Minnesota streams. Alkalinity is assumed to be an index to stream productivity, since carbonate and bicarbonate usually predominate in alkalinity and carbon is an essential algal nutrient. More productive streams are considered likely to yield more drift. In my study, however, a significant relationship between alkalinity and algae was not found among all the streams. Instead, a significant relationship was found between total phosphorus and stream algae, that explained nearly 60% of the variation in the chlorophyll a content of the suspended algae from stream-to-stream;When the streams were grouped into two types, clearwater (boreal) and brownwater (muskeg), the relationships between alkalinity and algae was positive and significant, as expected, for clearwater streams, but negative and non-significant for brownwater streams. Additionally, no significant relationships were found between algae and invertebrate drift for either stream type. It was concluded that allochthonous materials may be additional or substitute foods (besides algae) for invertebrates in those streams;Leaf litter imputs to clearwater streams in subarctic Alaska are small relative to those to temperate streams. However, utilization of leaf litter by aquatic organisms is known in these systems. Leaf litter imputs to brownwater streams are even less than to clearwater streams because trees are rare in muskeg. In brownwater streams the salts of organic acids such as tannins and lignins, which titrate as "false" (non-carbonate) alkalinity, may provide food to the invertebrates either directly as particulates or by providing a substrate for edible bacteria. This may help to explain the significant relationship between alkalinity and invertebrate drift among all the study streams;Previous studies that have noted the effects of stream flow on drift have each only considered a single stream. Those findings explain the effects of storm spates on drift. My study identified stream-to-stream effects of flow variation. I found that in two streams with the same discharge but with different average velocities the greater amount of drift would be produced in the stream with the greater average velocity. As such, drift transport seems analogous to sediment transport in streams. Therefore, the hypothesis that invertebrate drift is primarily the result of hydromechanical dislodgement was found acceptable.