Genetic studies have emphasized the contrast between North African and sub‐Saharan populations, but the particular affinities of the North African mtDNA pool to that of Europe, the Near East, and sub‐Saharan Africa have not previously been investigated. We have analysed 268 mtDNA control‐region sequences from various Northwest African populations including several Senegalese groups and compared these with the mtDNA database. We have identified a few mitochondrial motifs that are geographically specific and likely predate the distribution and diversification of modern language families in North and West Africa. A certain mtDNA motif (16172C, 16219G), previously found in Algerian Berbers at high frequency, is apparently omnipresent in Northwest Africa and may reflect regional continuity of more than 20000 years. The majority of the maternal ancestors of the Berbers must have come from Europe and the Near East since the Neolithic. The Mauritanians and West‐Saharans, in contrast, bear substantial though not dominant mtDNA affinity with sub‐Saharans.

Although the Canary Islands were settled by humans, possibly of Berber origin, as late as 2500 years ago, the precise course and numbers of early migrations to the archipelago remain controversial. We have therefore analysed mtDNA variation (HVS‐I as well as selected RFLP sites) in 300 individuals from the seven Canary Islands. The distribution and variation across the islands in a specific mtDNA clade of Northwest African ancestry suggest that there was one dominant initial settlement process that affected all the islands, from east to west. This indicates that a certain genetic affinity of present‐day Canary Islanders to Northwest African Berbers mainly stems from the autochthonous population rather than slaves captured on the neighbouring African coast. The slave trade after the European conquest left measurable, though minor, traces in the mtDNA pool of the Canary Islands, which in its majority testifies to the European immigration.

ADP-glucose pyrophoshorylase (ADP-Glc PPase) catalyzes the first and limiting step in starch biosynthesis. In plants, the enzyme is composed of two types of subunits (small and large) and is allosterically regulated by 3-phosphoglycerate and phosphate. The pattern of expression and sugar regulation of the six Arabidopsis thaliana ADP-Glc PPase-encoding genes (two small subunits, ApS1 and ApS2; and four large subunits, ApL1–ApL4) has been studied. Based on mRNA expression, ApS1 is the main small subunit or catalytic isoform responsible for ADP-Glc PPase activity in all tissues of the plant. Large subunits play a regulatory role, and the data presented define a clear functional distinction among them. ApL1 is the main large subunit in source tissues, whereas ApL3 and, to a lesser extent, ApL4 are the main isoforms present in sink tissues. Thus, in source tissues, ADP-Glc PPase would be finely regulated by the 3-phosphoglycerate/phosphate ratio, whereas in sink tissues, the enzyme would be dependent on the availability of substrates for starch synthesis. Sugar regulation of ADP-Glc PPase genes is restricted to ApL3 and ApL4 in leaves. Sugar induction of ApL3 and ApL4 transcription in leaves allows the establishment of heterotetramers less sensitive to the allosteric effectors, resembling the situation in sink tissues. The results presented on the expression pattern and sugar regulation allow us to propose a gene evolution model for the Arabidopsis ADP-Glc PPase gene family.