3 Results: 1.3 mm continuum images
and circumstellar masses

Our on-the-fly 1.3 mm continuum maps are presented in Figs. 1 and 9 (for Taurus sources), in Figs. 2a-c and 10a-d (for isolated globules), and in Figs. 2d,e and 10e-g (for YSOs in Perseus). (Figs. 9 and 10 are only available in electronic form at http://www.edpsciences.org.)

It is apparent from the maps that most sources are associated with spatially resolved 1.3 mm dust emission, as expected for "bona fide'' embedded protostars surrounded by extended, relatively massive circumstellar envelopes. Surprisingly enough, however, 10 of the 27 candidate protostars of our Taurus subsample are barely resolved and one (K04181+2655) is even not detected. In the following, these 11 Taurus embedded YSOs, as well as the isolated IRAS globule L260 (with similar 1.3 mm properties), will be called "peculiar Class I sources''.

The maps also demonstrate that none of the objects of our sample is truly isolated. Most, if not all, of them must have formed in groups (especially in Perseus). At least seven of our maps show evidence for diffuse cloud emission and reveal the presence of new starless dense cores/fragments in the vicinity of the targeted YSOs of Table 1 (see Col. 8 of Table 2).

Basic information extracted from these maps is provided in Table 2 for the YSO sources of Table 1, and in Table 3 for the dense cores without IRAS emission. Table 2 lists the adopted name (Col. 1), the SED class (Col. 2, determined from the $\mbox{$T_{\mbox{\tiny bol}}$ }$ value of Table 1), the observing mode and year of observation (Col. 3), the peak and integrated flux densities (Cols. 4 and 5), the estimated circumstellar mass within a radius of $4\,200$ AU (Col. 6), the degree of flux concentration ( $\mbox{$S^{\mbox{\tiny 11\hbox{$^{\prime\prime}$ }}}/S^{\mbox{\tiny 60\hbox{$^{\prime\prime}$ }}}$ }$ in Col. 7), and a short description of the source environment (Col. 8). Table 3 gives similar information for the new starless cores.

 

 

Table 2: Circumstellar fluxes and masses

Adopted	SED	Observing	$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~peak}}$(2)	$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~int}}$(3)	$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ (4)	Concentration	Source
source name	class	method (1)	(mJy/beam)	(mJy)	($M_\odot$)	$S^{\mbox{\tiny 11\hbox{$^{\prime\prime}$ }}}/S^{\mbox{\tiny 60\hbox{$^{\prime\prime}$ }}}$(5)	environment
L1489	I	6$\times$otf9699	$130\pm5$	150	0.03	U, 90%	L1489-NH3
M04108-A	II	2$\times$otf99	< 20	-	< 0.002	-	L1495N-NH3
M04108-B	I	2$\times$otf99	$39 \pm5$	39	0.008	U, 100%	L1495N-NH3
K04113	I	otf99	$410\pm40$	750	0.4	55%	diffuse cloud
Elias1	II	otf94	$270\pm35$	270	0.03	U, 100%
K04158	I	otf96	$110\pm5$	110	0.025	U, 100%
K04166	0/I	2$\times$otf9599	$180\pm8$	800	0.45	20%
K04169	I	2$\times$otf9599	$190\pm9$	730	0.40	30%	K04169-NW
K04181+2655	I	2$\times$otf9699	< 10	-	< 0.002	-	K04181+2654
K04181+2654	I	2$\times$otf9699	$50\pm8$	230	0.12	20%
IRAM04191	0	3$\times$otf9599	$110\pm7$	650	0.45 $^{\mbox{\tiny 12.5K}}$	20%	T04191
T04191	I	3$\times$otf9599	$110\pm7$	400	0.2	30%	IRAM04191
M04239	I	mpt93	$80\pm10$	170	$\sim$0.09	$\sim50\%$
M04248	I	2$\times$otf99	$60\pm7$	450	0.25	10%	B217-NH3
Z04260	I	2$\times$otf99	$105\pm10$	120	0.025	U, 90%
Haro6-10	I	2$\times$mpt93	$100\pm10$	200	$\sim$ $0.07^{\mbox{\tiny 20K}}$	$\sim$50%
Elias6	I	otf99	$31\pm8$	31	0.007	U, 100%
L1551-IRS5	I	4$\times$otf99	$1\,500\pm7$	$3\,400$	0.90 $^{\mbox{\tiny 25K}}$	40%	L1551-NE
HH30-IRS	I/II	otf99	$35\pm10$	35	0.008	U, 100%	HLTau, diffuse cloud
HLTau	II	otf99	$960\pm10$	$1\,200$	0.13	U, 80%	diffuse cloud
L1551-NE	I	2$\times$otf99	$850\pm10$	$1\,500$	0.55 $^{\mbox{\tiny 20K}}$	60%	IRS5, cloud fragments
M04295	I	otf96	$115\pm10$	115	0.025	U, 100%
GGTau	II	otf93	$665\pm10$	665	0.07	U, 100%
K04302	I	otf96	$180\pm10$	180	0.04	U, 100%
T04325	I	3$\times$otf99	$110\pm7$	520	0.3	20%	L1535-NW, cl. fragment
TMR1	I	otf99	$110\pm8$	440	0.25	25%	diffuse cloud
TMC1A	I	mpt93	$230\pm10$	450	$\sim$0.25	$\sim$50%
L1527	0	2$\times$otf95	$375\pm6$	$1\,500$	0.80	25%	cloud fragment
M04381	I	otf99	$70\pm9$	300	0.16	20%	diffuse cloud
TMC1C	I	otf99	$30\pm5$	30	0.007	U, 100%
M04489	I	otf99	$15\pm4$	$\lower.5ex\hbox{$\; \buildrel > \over \sim \;$ }15$	$\lower.5ex\hbox{$\; \buildrel > \over \sim \;$ }0.003$	U, 100%
B35	I	mpt93	$210\pm5$	820	$\sim$4.0	$\sim$30%
L260	I	otf94	$110\pm5$	110	0.03	U, 100%	diffuse cloud L260-NH3
L483-MM	0	otf93	$290\pm15$	$1\,100$	0.8	15%
L588	I	otf93	$295\pm5$	570	0.4	40%
L723-MM	0	otf94	$270\pm 20$	370	0.6	40%
B335	0	3$\times$otf9495	$270\pm 5$	780	0.9	20%
L1157-MM	0	otf94	$630\pm10$	630	2.0	40%
B361	I	otf96	$170\pm10$	190	0.4	50%	B361-NH3
L1262	I	2$\times$mpt94	$120\pm 15$	500	$\sim$0.35	$\sim$20%
L1448-NW	0	otf93	$560\pm25$	900	1.5	35%	L1448-N and cloud
L1448-N	0	otf93	$1\,400\pm5$	$2\,100$	3.5	40%	L1448-C, -NW and cloud
L1448-C	0	otf93	$620\pm15$	910	1.5	40%	L1448-N and cloud
NGC 1333-IRAS 2	0	otf94	$830\pm12$	875	1.2 $^{\mbox{\tiny 30K}}$	40%	cloud fragment
NGC 1333-IRAS 4A	0	otf93	$4\,100\pm20$	$4\,100$	4.5 $^{\mbox{\tiny 35K}}$	65%	IRAS 4B and cloud
NGC 1333-IRAS 4B	0	otf93	$1\,470\pm40$	$1\,500$	1.7 $^{\mbox{\tiny 35K}}$	U, 90%	IRAS 4A
IRAS 03282	0	otf95	$415\pm10$	590	1.4 $^{\mbox{\tiny 15K}}$	40%
HH211-MM	0	otf95	$395\pm15$	900	1.5	30%	cloud fragment
B5-IRS1	I	mpt93	$80\pm5$	200	$\sim$0.35	$\sim$40%

Notes: (1) Number, method and date of 1.3 mm observations: "otf'' = on-the-fly map; "mpt'' = multi-point map. (2) Peak flux in a 11 $\hbox{$^{\prime\prime}$ }$-12 $\hbox{$^{\prime\prime}$ }$ beam at source peak. Uncertainty corresponds to $1\sigma$ noise level of on-off and map measurements at peak position, upper limits are $3\sigma$. (3) Flux integrated over a $4\,200$ AU radius circle for on-the-fly maps, Gaussian integrated flux for multi-point maps. (4) Circumstellar mass within a $4\,200$ AU radius circle: approximate value derived from multi-point maps. Assumptions on dust opacity and dust temperature are given in Sect. 3.1. (5) Ratio of peak flux (in 11 $\hbox{$^{\prime\prime}$ }$-beam) to integrated flux in 60 $\hbox{$^{\prime\prime}$ }$ diameter circle. "U'' indicates unresolved sources.

Figure 1: Selection of on-the-fly 1.3 mm continuum maps obtained toward Taurus embedded YSOs (all smoothed to a $13\hbox {$^{\prime \prime }$ }$ effective beam). The positions of YSOs with weak emission and starless fragments are marked by stars and crosses, respectively. Contour levels and rms noise at map center are: a) 15 to 120 by 15 mJy/beam, $1\sigma \simeq 4$ mJy/beam; b) 35, 70, 105 mJy/beam, $1\sigma \simeq 9.5$ mJy/beam; c) 90 to 450 by 90 mJy/beam, $1\sigma \simeq 27$ mJy/beam; d) 25 to 200 by 25 mJy/beam, $1\sigma \simeq 6\$to 7 mJy/beam

Figure 1: continued. Contour levels and rms noise at map center are: e) 17 to 136 by 17 mJy/beam, $1\sigma \simeq 5$ mJy/beam; f) 17 to 68 by 17 mJy/beam, $1\sigma \simeq 5$ mJy/beam; g) 25 to 150 by 25 mJy/beam, then 200 to 800 by 200 mJy/beam and 1.2, 1.6 Jy/beam, $1\sigma \simeq 6\$to 9 mJy/beam; h) 40 to 200 by 40 mJy/beam and then 300, 600, 900 mJy/beam, $1\sigma \simeq 12$ mJy/beam; i) 17 to 187 by 17 mJy/beam, $1\sigma \simeq 5$ mJy/beam; j) 15 to 135 by 15 mJy/beam, $1\sigma \simeq 4.5$ mJy/beam

Figure 1: continued. Contour levels and rms noise at map center are: k) 40, 80, 120 mJy/beam, $1\sigma \simeq 11$ mJy/beam; l) 12 to 120 by 12 mJy/beam and 180 to 420 by 60 mJy/beam, $1\sigma \simeq 3.5$ mJy/beam; m) 25, 50, 75 mJy/beam, $1\sigma \simeq 7$ mJy/beam. The outflow axis and proposed companion of L1527 are marked with arrows and a filled square marker in l) (cf. MacLeod et al. 1994; Fuller et al. 1996, respectively). See complementary Fig. 9 within the on-line version of this paper

Figure 2: Same as Fig. 1 for several isolated IRAS globules ( a- c) and Perseus protostars ( d, e). Contour levels and rms noise at map center are: a) 12 to 96 by 12 mJy/beam and 150 to 300 by 50 mJy/beam, $1\sigma \simeq 3.5$ mJy/beam; b) 25 to 175 by 25 mJy/beam, $1\sigma \simeq 7.5$ mJy/beam; c) 40 to 120 by 40 mJy/beam and 200 to 600 by 80 mJy/beam, $1\sigma \simeq 11$ mJy/beam; d) 25 to 100 by 25 mJy/beam and 150 to 450 by 50 mJy/beam, $1\sigma \simeq 7.5$ mJy/beam; e) 30, 60, 90 mJy/beam and 150 to 500 by 50 mJy/beam, $1\sigma \simeq 8.5$ mJy/beam. The axis of the B335 outflow is indicated by arrows in a). See complementary Fig. 10 on-line

 

 

Table 3: Cold dense cores with no IRAS emission

Core	Coordinates		d	$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~peak}}$(2)	$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~int}}$(3)	$M^{\mbox{\tiny 4200~AU}}$(4)	$R_{\mbox{\tiny out}}$(5)	$M^{\mbox{\tiny\mbox{$R_{\mbox{\tiny out}}$ }}}$(6)
name (1)	$\alpha_{\rm 1950}$	$\delta_{\rm 1950}$	(pc)	(mJy/beam)	(mJy)	($M_\odot$)	(AU)	($M_\odot$)
L1489-NH3	04$^{\rm h}$01$^{\rm m}$45 $\hbox{$.\!\!^{\rm s}$ }$7	26 $\hbox{$^\circ$ }$11 $\hbox{$^\prime$ }$10 $\hbox{$^{\prime\prime}$ }$	140	$55\pm5$	510	1.0	$15\,000\pm 3\,000$	4.
L1495N-NH3	04$^{\rm h}$10$^{\rm m}$45 $\hbox{$.\!\!^{\rm s}$ }$0*	28 $\hbox{$^\circ$ }$05 $\hbox{$^\prime$ }$40 $\hbox{$^{\prime\prime}$ }$*	140	$(45\pm15)$*	$\sim$500	$\sim$1.0	$\lower.5ex\hbox{$\; \buildrel > \over \sim \;$ }14\,000$	>5.0*
K04169-NW	04$^{\rm h}$16$^{\rm m}$47 $\hbox{$.\!\!^{\rm s}$ }$4	27 $\hbox{$^\circ$ }$04 $\hbox{$^\prime$ }$05 $\hbox{$^{\prime\prime}$ }$	140	$50\pm15$	340	0.7	$\lower.5ex\hbox{$\; \buildrel > \over \sim \;$ }14\,000$	2.5
B217-NH3	04$^{\rm h}$24$^{\rm m}$45 $\hbox{$.\!\!^{\rm s}$ }$8	26 $\hbox{$^\circ$ }$11 $\hbox{$^\prime$ }$30 $\hbox{$^{\prime\prime}$ }$	140	$55\pm10$	560	1.1	$14\,000\pm 1\,500$	5.5
L1535-NE	04$^{\rm h}$32$^{\rm m}$36 $\hbox{$.\!\!^{\rm s}$ }$3	24 $\hbox{$^\circ$ }$03 $\hbox{$^\prime$ }$14 $\hbox{$^{\prime\prime}$ }$	140	$100\pm7$	850	1.7	$\lower.5ex\hbox{$\; \buildrel > \over \sim \;$ }21\,000$	7.5
B361-NH3	21$^{\rm h}$10$^{\rm m}$33 $\hbox{$.\!\!^{\rm s}$ }$7	47 $\hbox{$^\circ$ }$12 $\hbox{$^\prime$ }$46 $\hbox{$^{\prime\prime}$ }$	350	$90\pm10$	130	1.6	$35\,000\pm3\,500$	25.

Notes: (1)  The "NH3'' suffix refers to Benson & Myers (1989); "NE'' and "NW" correspond to new dust cores. (2) to (4)  Same as in Table 2. (5)  Estimated outer radius. (6)  Total mass within $R_{\mbox{\tiny out}}$. ^*  The characteristics of L1495N-NH3 are more uncertain as this source was incompletely mapped.

3.1 Circumstellar mass estimates

Since dust continuum emission is largely optically thin at 1.3 mm, our bolometer maps should primarily reflect the dust column density distribution within the observed circumstellar envelopes. If the dust properties and the gas-to-dust ratio are uniform, the 1.3 mm flux density $S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~beam}}$ measured in a single beam should even be directly proportional to the total (gas + dust) beam-averaged column density along the line of sight: $\mbox{$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~beam}}$ }\, \propto \: \k13 \; \B13 \: \times <N\h2>\mbox{$_{\mbox{\tiny beam}}$ }$, where 13 is the dust opacity per unit mass column density at $\lambda = 1.3$ mm, and 13 is the Planck function for a dust temperature $T_{\mbox{\tiny dust}}$ (see, e.g., MAN98 for a complete equation). Accordingly, the integrated flux density is directly related to the total mass of emitting material. Column 6 of Table 2 gives the circumstellar mass contained within a radius of $4\,200$AU of each object, derived from the 1.3 mm flux density integrated over a $30\hbox{$^{\prime\prime}$ }\times(d/140~{\rm pc})^{-1}$ radius circle, $\mbox{$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~int}}$ }$ (listed in Col. 5), as follows:

 

$$\mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ } \frac{\mbox{$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~int}}$ }\; d^{2}}{\k13 \; \B13}$$ =

$$\simeq 5.3~10^{-3}\ \mbox{$M_\odot$ }\times \left(\frac {\mbox{$S_{\mbox... ...\tiny ~int}}$ }}{\mbox{10~mJy}}\right) \left(\frac {d}{\mbox{140~pc}} \right)^2$$

$$\times \left(\frac {\k13}{0.01\,\mbox{$\mbox{cm}^{2} \, \mbox{g}^... ...^{-1} \left(\frac {<\mbox{$T_{\mbox{\tiny dust}}$ }>}{15~\mbox{K}}\right)^{-1}.$$ (1)

For consistency with previous work (e.g. AM94, AWM96, MAN98), we adopted dust opacities per unit (gas + dust) mass column density of $\k13=0.005\;\mbox{$\mbox{cm}^{2} \, \mbox{g}^{-1}$ }$, $\k13=0.01\;\mbox{$\mbox{cm}^{2} \, \mbox{g}^{-1}$ }$, and $\k13=0.02~\mbox{$\mbox{cm}^{2} \, \mbox{g}^{-1}$ }$ for starless cores, envelopes of Class 0/Class I protostars, and disks around Class II sources, respectively. These opacities correspond to the values recommended by Henning et al. (1995) assuming a gas-to-dust mass ratio of 100. Considering the likely evolution of the gas-to-dust mass ratio (cf. Ciolek & Mouschovias 1996) and of the dust properties themselves (Henning et al. 1995), the adopted opacities are believed to be uncertain by a factor of $\lower.5ex\hbox{$\; \buildrel > \over \sim \;$ } 2$ on either side of the quoted values. This is in agreement with models of dust in protostellar cores (e.g. Ossenkopf & Henning 1994), laboratory measurements (Agladze et al. 1996), and recent cross-comparisons with dust extinction observations (Kramer et al. 1999).

The value $<\mbox{$T_{\mbox{\tiny dust}}$ }>$ used for the dust temperature in Eq. (1) corresponds to a mass-weighted average calculated up to a radius of $4\,200$ AU (i.e., an angular radius of $30\hbox{$^{\prime\prime}$ }\times(d/140~{\rm pc})^{-1}$) and for a centrally-heated sphere with a $\rho (r) \propto r^{-2}$ density gradient ( $<\mbox{$T_{\mbox{\tiny dust}}$ }>$ would decrease by only $14\%$ if a $\rho(r) \propto r^{-1.5}$ density gradient were adopted instead). In the case of unresolved sources, a similar weighted average temperature was calculated, but only up a radius of $800~{\rm AU}\times (d/140~{\rm pc})$ (i.e. HPBW $/2 \simeq 5.5\hbox{$^{\prime\prime}$ }$). The radial temperature profile discussed in Sect. 4.3 below was assumed. In practice, this means that we adopted $\mbox{$<\mbox{$T_{\mbox{\tiny dust}}$ }>_{4\,200\;\rm AU}$ }= 15\ \mbox{K}$ for the envelopes surrounding the low-luminosity protostars of Taurus and $\mbox{$<\mbox{$T_{\mbox{\tiny dust}}$ }>_{4\,200\;\rm AU}$ }= 20\ \mbox{K}$ for the more luminous isolated IRAS globules and Perseus protostars. We assumed $<T_{\rm dust}> = 10$ K for the starless cores and $<T_{\rm dust}>_{800~\rm AU} = 30$ K for the unresolved YSO sources.

The reason why we chose a fiducial radius of $4\,200$ AU for the circumstellar mass in Eq. (1) and Table 2 is that this roughly corresponds 1) to the head of the expansion wave for a 10⁵-yr old protostar in the inside-out collapse scenario if the sound speed is $\mbox{$a_{\mbox{\tiny s}}$ }\sim 0.2$ km s^-1 (i.e. $\mbox{$T_{\mbox{\tiny cloud}}$ }\sim 10$ K), and 2) to an angular radius of 30 $^{\prime\prime}$ at the distance of Taurus which is comparable to the radius of the IRAS and ISO beams at $60\ \mu$m.

For the entire sample of 27 Taurus candidate protostars (including peculiar Class Is), the median mass enclosed within 4200 is $\overline{\mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ }}(\r4200)\sim 0.074~\mbox{$M_\odot$ }$. This value is not much larger than the typical disk mass found for classical T Tauri stars (Beckwith et al. 1990) or embedded Class II sources (AM94). In contrast, the 16 Taurus sources of Table 2 with spatially resolved emission are bona-fide protostars with significantly more massive circumstellar structures (envelopes/disks): $\mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ }\sim 0.07-0.9\ \mbox{$M_\odot$ }$, corresponding to a median mass $\overline{\mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ }}(\r4200) \sim 0.28\ \mbox{$M_\odot$ }$. The bona-fide protostars observed in Perseus and Bok globules have even larger circumstellar masses: $\mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ }\sim 0.3{-}5\ \mbox{$M_\odot$ }$ and $\mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ }\sim 0.3{-}4\ \mbox{$M_\odot$ }$, respectively.

3.2 Envelope versus disk mass

A priori, the integrated fluxes listed in Table 2 include contributions from both the protostellar envelope and the circumstellar disk: $\mbox{$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~int}}$ }= S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~env}} + S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~disk}}$. The disk contribution must be properly assessed if an accurate estimate of the envelope mass is desired.

Predictions made in the framework of the standard theory of isolated protostars suggest that the millimeter continuum flux should be dominated by emission from the envelope rather than from the disk, when observed at the resolution of the IRAM 30 m telescope (e.g. Terebey et al. 1993; Galli 1995). In agreement with these predictions, the median flux concentration $\mbox{$S^{\mbox{\tiny 11\hbox{$^{\prime\prime}$ }}}/S^{\mbox{\tiny 60\hbox{$^{\prime\prime}$ }}}$ }\sim 30\%$ estimated for the bona-fide protostars of our sample suggests that $\lower.5ex\hbox{$\; \buildrel > \over \sim \;$ }70\%$ of the integrated flux density $\mbox{$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~int}}$ }$ arises from the envelope. Millimeter interferometric observations by, e.g., Hogerheijde et al. (1997) and Motte et al. (2001) provide estimates of the disk component $S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~disk}}$ for several bona-fide protostars in Taurus, and confirm that the integrated flux density measured at the 30 m telescope arises primarily from the envelope: $S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~disk}}/\mbox{$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~int}}$ }(\r4200) \lower.5ex\hbox{$\; \buildrel < \over \sim \;$ }10\%$. A similar conclusion holds in Perseus for the Class 0 protostars NGC 1333-IRAS 4A and NGC 1333-IRAS 4B which, albeit barely resolved at the $11\hbox{$^{\prime\prime}$ }$- $13\hbox {$^{\prime \prime }$ }$ resolution of the 30 m and JCMT telescopes (cf. Table 2 and Sandell et al. 1991), are $\sim$8 times stronger than when observed with the CSO-JCMT interferometer (Lay et al. 1995). More generally, the disk contribution is estimated to be $\lower.5ex\hbox{$\; \buildrel < \over \sim \;$ }10\%$ for 8 of the 9 Perseus protostars of our sample (Motte et al. 2001; Looney et al. 2000).

The above discussion suggests that $\mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ }$ should be a good approximation of the envelope mass within $4\,200$AU for all the "self-embedded'' sources, even when the exact disk contribution is unknown. In the following, we will thus assume $\mbox{$M_{\mbox{\tiny env}}^{\mbox{\tiny 4200~AU}}$ }\sim \mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ }$ for all the protostars with spatially resolved envelopes (see Col. 7 of Table 2).

By contrast, the disk can sometimes contribute a large fraction of the peak flux density $\mbox{$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~peak}}$ }$ measured in an 11 $\hbox{$^{\prime\prime}$ }$ beam. For the "peculiar'', unresolved Class I sources of our sample, most of the circumstellar material may be in a disk, with $S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~disk}}/\mbox{$S_{\mbox{\tiny 1.3~mm}}^{\mbox{\tiny ~peak}}$ }\lower.5ex\hbox{$\; \buildrel < \over \sim \;$ }100\%$ (see discussion in Sect. 5.2.3 below). In these cases, the masses listed in Table 2 provide only upper limits to the actual envelope masses: $\mbox{$M_{\mbox{\tiny env}}^{\mbox{\tiny 4200~AU}}$ }<\mbox{$M_{\mbox{\tiny c$\star$ }}^{\mbox{\tiny 4200~AU}}$ }$.