This paper uses multipoint measurements to analyse solar wind magnetic holes crossing the terrestrial bow shock and entering into the magnetosheath. Data from the Cluster four-spacecraft constellation have been used to observe density holes upstream of the bow shock and also shortly after, downstream in the magnetosheath. The analysis of 26 such magnetic hole observations shows that the temporal scale size and the magnetic field rotation are very similar for the upstream and downstream observations. This is consistent with the interpretation that magnetic holes are of solar wind origin. 

The analysis is careful, the results are original and the paper is well written. The paper should be accepted for publication after a minor revision.

Specific Comments:

Magnetic holes are part of the “zoo” of several transient structures that have been observed upstream of the Earth’s bow shock. These include: (i) hot flow anomalies, which are also characterised by a depressed magnetic field but are filled with hot plasma flowing in a direction significantly deflected from the solar wind velocity vector (e.g. Lucek et al., J. Geophys. Res., doi:10.1029/2003JA010016, 2004); (ii) hot diamagnetic cavities, where the depressed magnetic field within the cavity is flanked by strong enhancements (e.g. Thomsen et al., J. Geophys. Res., doi:10.1029/JA091iA03p02961, 1986); (iii) foreshock cavities, where temperature and pressure inside are only slightly greater than in the ambient solar wind (Sibeck et al., J. Geophys. Res., doi:10.1029/2001JA007539, 2002); and (iv) solar wind density holes, characterised by a strong plasma density depletion within them, flanked by density overshoots and compressed magnetic field (Parks et al., Phys. of Plasmas Lett., 13, 050701, 2006). What is missing from the present paper is putting magnetic holes into context and comparing them with these other transient upstream structures. This is particularly useful for their relationship with density holes, since as shown by Parks et al. (2006) density holes are accompanied by magnetic holes of nearly the same shape.

The manuscript presents an analysis of multipoint magnetic field and plasma measurements provided by Cluster in order to study magnetic holes in the solar wind and the terrestrial magnetosheath. In particular, the authors have identified 26 magnetic holes upstream of the bow shock and detected them shortly after, downstream in the magnetosheath. This study also presents a correlation between upstream and downstream observations of some of the magnetic holes' properties. The authors find that the temporal scale size and magnetic field rotation across the magnetic holes are very similar for the upstream and downstream observations, while the depth of the magnetic holes varies more. The results are consistent with the interpretation that magnetic holes are of solar wind origin, as suggested in previous works.

The manuscript presents a comprehensive and interesting analysis, reports new findings, and is well-written. I have only minor comments for the authors’ consideration.

Lines 121-123: This seems to be a suggestion rather than a conclusion. I suggest rephrasing this sentence.

Lines 130-134: ‘We will define an event as a magnetic hole event if a localized magnetic field decrease is below -0.5 in either the solar wind or the magnetosheath region, and there is a similar structure with a decrease of at least -0.4 in the ’complementary’ region (in this case the solar wind.)’ These seem to be arbitrary numbers. Is there a physical reason for this choice? Do the results presented in this work vary significantly if the authors consider different values for the magnetic field decrease in both regions?

Figure 3: How do the authors distinguish between magnetic holes (particularly rotational magnetic holes) in the solar wind and current sheet crossings (e.g, heliospheric current sheet crossings)?

Line 246: The authors stated they identified 26 events of interest. I think it would be worth adding the amount of data that has been analyzed to be able to find them. How often magnetic holes are observed by Cluster?