Since the invention of arterial spin labeling (ASL) it has been acknowledged that ASL does not allow reliable detection of a white matter (WM) perfusion signal. However, recent developments such as pseudo-continuous labeling and background suppression have improved the quality. The goal of this research was to study the ability of these newer ASL sequences to detect WM perfusion signal. Background suppressed pseudo-continuous ASL was implemented at 3T with multislice 2D readout after 1525 ms. In five volunteers it was shown that 10 min scanning resulted in significant perfusion signal in 70% of WM voxels. Increasing the labeling and delay time did not lead to a higher percentage. In 27 normal volunteers it was found that 35 averages are necessary to detect significant WM signal, but 150 averages are needed to detect signal in the deep WM. Finally, it was shown in a patient with a cerebral arteriovenous malformation that pseudo-continuous ASL enabled the depiction of hypointense WM perfusion signal, although dynamic susceptibility contrast MRI showed that this region was merely showing delayed arrival of contrast agent than hypoperfusion. It can be concluded that, except within the deep WM, ASL is sensitive enough to detect WM perfusion signal and perfusion deficits.